University of Southern California Dissertations and Theses

Scanning using a smart phone for heritage conservation: a case study using the Reunion House

(USC Thesis Other)

Scanning using a smart phone for heritage conservation: a case study using the Reunion House

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

SCANNING USING A SMART PHONE FOR HERITAGE CONSERVATION:
A Case Study Using the Reunion House
by
Ye Hong

A Thesis Presented to the
FACULTY OF THE USC
SCHOOL OF ARCHITECTURE
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements of Degree
MASTER OF BUILDING SCIENCE AND HERITAGE CONSERVATION

MAY 2023

`
ii

Acknowledgements
I would like to express my deepest gratitude to all those who have contributed to the completion
of my thesis project. Without their support, guidance, and encouragement, this achievement would not
have been possible.

Firstly, I would like to extend my sincere appreciation to my Thesis Chair, Karen Kensek, and
committee members, Peyton Hall and Trudi Sandmeier, for their invaluable guidance, feedback, and
support throughout my research. Their expertise and advice have been instrumental in shaping my work
and ensuring its quality.

I am also grateful to Sian Winship, Raymond Neutra, and Neutra Institute for providing me with
the opportunity to scan and research the Reunion House. Their support and assistance were vital to the
success of my project, and I am grateful for their generosity.

I would like to acknowledge Alan White and Christopher Gray for their invaluable advice and
knowledge, which have contributed significantly to the development of my research. Their expertise has
been an inspiration to me, and I am grateful for their willingness to share their insights with me.
I would also like to express my appreciation to my peers for their support and constructive feedback,
which have helped me improve my work. Their encouragement and camaraderie have been essential to
my success.

Finally, I would like to acknowledge my family for their unwavering support and encouragement
throughout this project. Their love and encouragement have been a constant source of inspiration for me,
and I am grateful for their presence in my life.

Once again, I would like to express my sincere appreciation to everyone who has contributed to
the completion of my thesis project. Their support, guidance, and encouragement have been invaluable,
and I am grateful for their contributions to my academic journey.

`
iii

TABLE OF CONTENTS

ACKNOWLEDGEMENT ............................................................................................................................................ ii
LIST OF TABLE ..........................................................................................................................................................vi
LIST OF FIGURE ...................................................................................................................................................... vii
ABSTRACT .................................................................................................................................................................xi
CHAPTER 1 INTRODUCTION ................................................................................................................................... 1
1.1 HERITAGE CONSERVATION AND TECHNOLOGY ................................................................................................... 1
1.1.1 What is Heritage Conservation? .................................................................................................................. 2
1.1.2 What do we conserve? ................................................................................................................................ 2
1.1.3 How do we conserve? ................................................................................................................................. 3
1.1.4 Heritage Conservation and technology – technology for Heritage Conservation ....................................... 6
1.2 DIGITAL DOCUMENTATION .................................................................................................................................. 6
1.2.1 2D documentation ....................................................................................................................................... 7
1.2.2 360-degree photographs ............................................................................................................................ 11
1.2.2 3D documentation using scanners ............................................................................................................. 14
1.3 ACCURACY VERSUS PRECISION .......................................................................................................................... 24
1.3.2 Accuracy of scanners ................................................................................................................................ 25
1.3.3 Point Cloud data processing software ....................................................................................................... 27
1.4 RICHARD NEUTRA AND REUNION HOUSE AS A CASE STUDY .............................................................................. 28
1.4.1 A Short biography of Richard and Dion Neutra ........................................................................................ 28
1.4.2 Reunion House .......................................................................................................................................... 29
1.5 CONCLUSION ..................................................................................................................................................... 30
CHAPTER 2 LITERATURE REVIEW ...................................................................................................................... 34
2.1 SCANNING FOR NON-HERITAGE CONSERVATION PURPOSES ............................................................................... 34
2.1.1 Triangulation scanners application ............................................................................................................ 34
2.1.2 LiDAR scanners application ..................................................................................................................... 37
2.1.3 Phase-Comparison scanners application ................................................................................................... 41
2.2 SCANNING HERITAGE ......................................................................................................................................... 42
2.2.1 Panoramic photo application in Heritage Conservation ............................................................................ 43
2.2.2 Triangulation scanners application in Heritage Conservation ................................................................... 44
2.2.2.1 Photogrammetry application in Heritage Conservation ....................................................................................... 44
2.2.2.2 Structured Light application in Heritage Conservation ....................................................................................... 46
2.2.3 Light Detecting and Ranging (LiDAR) application in Heritage Conservation ......................................... 48
2.2.4 Phase scanners application in Heritage Conservation ............................................................................... 50
2.2.5 360 Degree photograph application in Heritage Conservation.................................................................. 51
2.3 STUDIES USING A SMARTPHONE FOR SCANNING ................................................................................................. 53
2.4 EXISTING RESEARCH ON RICHARD AND DION NEUTRA’S REUNION HOUSE ....................................................... 54
2.4.1 Reunion House developmental history ...................................................................................................... 55
2.4.2 Reunion House Character Defining Features ............................................................................................ 55
2.4.3 Documentation techniques used previously on Reunion House................................................................ 58
2.5 SUMMARY .......................................................................................................................................................... 60
CHAPTER 3 METHODOLOGY ................................................................................................................................ 62
`
iv

3.1 IPHONE 13 PRO AS SCANNING DEVICE ................................................................................................................ 64
3.1.1 Scanning app: SiteScape and Matterport ................................................................................................... 65
3.2 HP SURVEY WITH SMARTPHONE ........................................................................................................................ 67
3.2.1 Software selection – SiteScape ................................................................................................................. 68
3.2.2 Data acquisition process ............................................................................................................................ 68
3.2.3 CloudCompare: merging multiple scans ................................................................................................... 69
3.2.4 Smartphone acquisition qualification ........................................................................................................ 70
3.3 HC DOCUMENTATION WITH SMARTPHONE ......................................................................................................... 71
3.3.1 Software selection – SiteScape ................................................................................................................. 71
3.3.2 Data acquisition process ............................................................................................................................ 72
3.3.3 Using CloudCompare processing data: increasing precision .................................................................... 72
3.3.4 Smartphone acquisition method qualification ........................................................................................... 73
3.4 MONITORING ..................................................................................................................................................... 74
3.4.1 Software selection – SiteScape ................................................................................................................. 74
3.4.2 Acquisition for HP monitoring .................................................................................................................. 75
3.4.3 Using CloudCompare processing data: increasing accuracy ..................................................................... 75
3.4.4 Smartphone acquisition qualification ........................................................................................................ 76
3.5 EDUCATION ....................................................................................................................................................... 76
3.5.1 Software selection – Matterport ................................................................................................................ 77
3.5.2 Data acquisition process ............................................................................................................................ 77
3.5.3 Using Matterport webpage processing scanned data ................................................................................. 78
3.5.4 Smartphone acquisition qualification ........................................................................................................ 79
3.6 SUMMARY .......................................................................................................................................................... 79
CHAPTER 4 TEST SCANS USING THE SMARTPHONE ...................................................................................... 81
4.1 SMARTPHONE SCANNING OVERVIEW AND DIRECTIONS FOR DOCUMENTATION ................................................... 84
4.1.1 Data acquisition ......................................................................................................................................... 85
4.1.2 Smartphone scanned data usage ................................................................................................................ 95
4.2 SMARTPHONE SCANNING OVERVIEW AND DIRECTIONS FOR SITE SURVEY .......................................................... 96
4.2.1 Data acquisition ......................................................................................................................................... 96
4.2.2 Data processing ......................................................................................................................................... 97
4.3 SMARTPHONE SCANNING OVERVIEW AND DIRECTIONS FOR MONITORING ........................................................ 119
4.3.1 Data acquisition ....................................................................................................................................... 119
4.3.2 Data processing ....................................................................................................................................... 121
4.4 SMARTPHONE SCANNING OVERVIEW AND DIRECTIONS FOR EDUCATION ......................................................... 131
4.4.1 Data acquisition ....................................................................................................................................... 132
4.4.2 Data processing ....................................................................................................................................... 136
4.5 SUMMARY ........................................................................................................................................................ 145
Chapter 5 Case Study Smartphone Scan Result ............................................................................................................ 148
5.1 EVALUATION STANDARD ................................................................................................................................. 149
5.2 SCANNED DATA RESULTS ................................................................................................................................. 151
5.2.1 Smartphone scanned data results ............................................................................................................. 164
5.2.2 Professional scanner scanned data results ............................................................................................... 165
5.3 RESULT COMPARISON BETWEEN IPHONE AND LEICA RTC 360 ........................................................................ 166
5.3.1 Measurements and errors ........................................................................................................................ 166
5.3.2 Visual comparison ................................................................................................................................... 171
5.4 QUALIFICATION ............................................................................................................................................... 174
5.5 SUMMARY ........................................................................................................................................................ 179
CHAPTER 6 QUALIFICATION .............................................................................................................................. 181
6.1 DOCUMENTATION ............................................................................................................................................ 183
6.2 SURVEY ........................................................................................................................................................... 190
6.3 MONITORING ................................................................................................................................................... 194
6.4 EDUCATION ..................................................................................................................................................... 195
`
v

6.5 CONCLUSION ................................................................................................................................................... 199
CHAPTER 7 CONCLUSION ................................................................................................................................... 201
7.1 SCANNING HERITAGE SITES WITH SMARTPHONES............................................................................................. 201
7.2 FUTURE WORK ................................................................................................................................................. 213
7.2.1 Improvements .......................................................................................................................................... 213
7.2.2 Other topic areas ..................................................................................................................................... 216
7.2.3 Future work ............................................................................................................................................. 218
7.3 SUMMARY ........................................................................................................................................................ 219

BIBLIOGRAPHY ......................................................................................................................................................................... 221
APPENDIX A ............................................................................................................................................................... 231

`
vi

List of Table
Table 1-1: 3D accuracy of 3D scanners ............................................................................................... 26
Table 1-2: Point clouds processing software information ................................................................... 27
Table 2-2: Scanning focus of each heritage conservation tasks .......................................................... 60
Table 3-1: Scanning focus for case study at Reunion House ............................................................... 72
Table 4-1: Test scan tasks of Hoose Library for heritage conservation purposes. .............................. 83
Table 5-1: Architectural Scales for drawing ...................................................................................... 150
Table 5-2: Engineering and map scales ............................................................................................. 151
Table 5-3: iPhone scanned high density point clouds measurements ................................................ 164
Table 5-4: Merged high density scans by iPhone measurements ...................................................... 164
Table 5-5: iPhone scanned low density point clouds measurements ................................................. 164
Table 5-6: Merged low density scans by iPhone measurements ....................................................... 165
Table 5-7: Merged floorplan measurements ...................................................................................... 165
Table 5-8: Leica RTC 360 scan measurements ................................................................................. 165
Table 5-9: Room width measurements comparison in section 1 ....................................................... 167
Table 5-10: Room height measurements comparison in section 2 .................................................... 167
Table 5-11: Bedroom width 2 measurements comparison in section 2 ............................................. 168
Table 5-12: Room height measurements comparison in section 3 .................................................... 169
Table 5-13: Desk height measurements comparison in Section 3 ..................................................... 169
Table 5-14: Point cloud thickness comparison in section 2 and 3 ..................................................... 170
Table 5-15: Smartphone scan low density mode error rate ............................................................... 175
Table 5-16: Smartphone scan low density mode merged error rate .................................................. 176
Table 5-17: Smartphone scan high density mode error rate .............................................................. 176
Table 5-18: Smartphone scan high density mode merged error rate ................................................. 177
Table 5-19: Smartphone scans merged floor plan error rate .............................................................. 177
Table 6-1: Architectural Scales for drawing ...................................................................................... 184
Table 7-1: Result in number of total points from point cloud from overlapping............................... 209
Table B-1 SiteScape test scan number of points ................................................................................ 240
Table B-2 Results of data processing for HC documentation............................................................ 240

`
vii

List of Figure
Figure 1-1 Library of Congress HABS Online Database Digital Collection ······························ 7
Figure 1-2: Measured drawing example ······································································· 9
Figure 1-3: Film photography of Reunion House living room ············································· 11
Figure 1-4: Raw 360-degree photograph ····································································· 12
Figure 1-5: 3D photograph ····················································································· 12
Figure 1-6: Google street view of Forbidden City ··························································· 13
Figure 1-7: Michael White Adobe virtual tour ······························································· 14
Figure 1-8: Point Clouds, Control Mesh, Result visualization of a rabbit ································ 16
Figure 1-9: Conegliano Italian Synagogue - Point Cloud ·················································· 17
Figure 1-10: Same point cloud data with different processing in reconstructing surface ··············· 18
Figure 1-11: Working principle of triangulation scanners ·················································· 19
Figure 1-12: Working principle of Structured Light scanners ············································· 20
Figure 1-13: Working principle of photogrammetry ························································ 21
Figure 1-14: Principles of laser scanner data acquisition, showing the example of TLS ··············· 22
Figure 1-15: Precision VS Accuracy ········································································· 25
Figure 1-16: Reunion house, Silver Lake, Los Angeles, Calif., 1951 ····································· 30
Figure 1-17: Heritage conservation task diagram ··························································· 31
Figure 1-18: Smartphone Scanning Methods ································································ 33
Figure 2-1: Structured Light Scanner ········································································· 35
Figure 2-2: The three major processing steps of PCs ······················································· 36
Figure 2-2 : Faro Focus Laser Scanner ······································································· 38
Figure 2-3: LiDAR drone concept ············································································· 39
Figure 2-4: iPad mounted on tripod, testing static configuration ·········································· 40
Figure 2-5: Phase Comparison Scanner ······································································ 41
Figure 2-6: Phase-comparison scanner scanned image ····················································· 42
Figure 2-7 Panoramic photographs of the Colosseum in Rome ··········································· 43
Figure 2-8: Orthoimage of mural painting by photogrammetry ··········································· 45
Figure 2-9 Photogrammetry from crowdsourced photography ············································ 46
Figure 2-10: Structured light scanning on the Minerva case study ········································ 47
Figure 2-11: TSL scanned data in Autocad measuring deformation of front wall ······················· 49
Figure 2-12: 360 degree panoramic photographs linked to BIM for heritage conservation ············ 51
Figure 2-4: Michael White Adobe presented by AQYER ················································· 52
Figure 2-12: Data comparison with TLS, DSLR, and SSL (smartphone) ································ 54
Figure 2-14: Smartphone scanning technologies match with heritage conservation goals ············· 61
Figure 3-1 Proposed methodology ············································································ 63
Figure 3-2 Using smartphone scan a cultural heritage ······················································ 65
Figure 3-3: SiteScape iOS app Figure 3-4: SiteScape parameter ··································· 66
Figure 3-5: Matterport Capture iOS app ······································································ 67
Figure 3-6: CloudCompare interface ·········································································· 70
Figure 3-7: Matterport Capture interface on iPhone ························································ 78
Figure 3-8: Matterport webpage editing interface ··························································· 79
Figure 4-1: Chapter 4 content diagram ······································································· 82
Figures 4-2 and 4-3: SiteScape register interface ···························································· 85
Figure 4-4: Hoose Library of Philosophy ···································································· 86
Figure 4-5, 4-6, and 4-7: SiteScape setting button and scan setting ······································· 87
Figures 4-9 and 4-10: SiteScape operation interface and camera view adjust ··························· 88
Figure 4-11 and 4-12: SiteScape pause and resume scanning button ····································· 89
Figure 4-13, 4-14, and 4-15: SiteScape post scan exporting and synchronizing ························· 90
Figure 4-16 and 4-17: SiteScape operation interface and camera view adjust ··························· 91
`
viii

Figure 4-18: The smartphone device on tripod with an angle facing down ······························ 92
Figure 4-19 and 4-20: Tripod angled up and down ························································· 92
Figures 4-21, 4-22, and 4-23: SiteScape library and renaming file········································ 93
Figures 4-24 and 4-25: Zoom in and out scan of Hoose Library of Philosophy on SiteScape ········· 94
Figure 4-26: Double layer of points in scan 2 ······························································· 94
Figures 4-27, 4-28, and 4-29: SiteScape post scan exporting and synchronizing ······················· 95
Figure 4-30: Heritage conservation survey by smartphone workflow diagram ·························· 96
Figure 4-31: SiteScape webpage log in ······································································· 98
Figure 4-32: SiteScape project database ······································································ 99
Figure 4-33: SiteScape measurement ········································································· 99
Figure 4-34: SiteScape floorplan ············································································ 100
Figure 4-35: SiteScape change point size ·································································· 100
Figure 4-36: Download options from SiteScape webpage ··············································· 101
Figure 4-37: Open file in CloudCompare ·································································· 101
Figure 4-38: Select the cloud in DB Tree window ························································ 102
Figure 4-39: SOR (Statistical Outlier Removal) setup ···················································· 103
Figures 4-40 and 4-41: Compute geometric feature and settings ········································ 103
Figures 4-42, 4-43, and 4-44: Filter by value, settings, and change color ······························ 104
Figure 4-45: CloudCompare Segment tool ································································· 105
Figure 4-46: Segment tool bar ··············································································· 105
Figures 4-47: Delete unwanted part ········································································· 106
Figures 4-48 and 4-49: Before and after segmenting ····················································· 106
Figures 4-50, 4-51, and 4-52: Point clouds automatically registered and merged ····················· 108
Figures 4-53 and 4-54: Select all point clouds and merge ················································ 109
Figures 4-55 and 4-56: After merge view in scalar field and how to change to RGB color ·········· 110
Figures 4-57 and 4-58: Example of mis-registration of point clouds ··································· 111
Figures 4-59, 4-60, 4-61, and 4-62: Translate/rotation setting, and manual alignment process ····· 112
Figure 4-63: Move to-be-aligned point cloud away from reference point cloud ······················ 113
Figures 4-64, 4-65, and 4-66: Align two clouds by picking four points tool ··························· 114
Figures 4-67 and 4-68: Pick equivalent points on both to-be-aligned and reference entities ········ 115
Figure 4-69: Merging result in scalar field ································································· 116
Figure 4-70: Merged library scans from sideview ························································· 117
Figures 4-71 and 4-72: Creating a slice of structure with segment tooln ······························· 118
Figure 4-73: Monitoring data process work flow ·························································· 121
Figure 4-74: Open file in CloudCompare ·································································· 122
Figure 4-75: Scan 3 point cloud in CloudCompare ······················································· 123
Figure 4-76: Scan 4 point cloud in CloudCompare ······················································· 124
Figure 4-77: Finely registers already(roughly) aligned entities (clouds or meshes) tool ············· 124
Figure 4-78: Cloud registration window setting ··························································· 125
Figure 4-79: Result of aligning scan 3 and scan 4 ························································ 126
Figure 4-80, Scan 4 as Reference and scan 3 as Compared in CloudCompare ························ 127
Figure 4-81: Distance computation parameter setup ······················································ 128
Figure 4-82, Approximate distance and Histogram for scan 3 and scan 4. ···························· 128
Figure 4-83, Cloud-to-cloud distance visualization from CloudCompare for scan 3 and scan 4. ··· 129
Figure 4-84, Saturation display range. ······································································ 130
Figure 4-85: Scan 3 and Scan 4 distance with saturation 1 inch ········································· 130
Figure 4-86: CloudCompare distance computation working principle ································· 131
Figure 4-87 and 4-88: Matterport Capture register interface ············································· 133
Figure 4-89: New job address information ································································· 134
Figures 4-90 and 4-91: Scan interface and parameter selection ········································· 134
Figures 4-92 and 4-93: Scan process ········································································ 135
`
ix

Figure 4-94: Trim and add window interface ······························································ 136
Figure 4-95: Matterport webpage ··········································································· 137
Figure 4-96: Matterport sign in ·············································································· 138
Figures 4-97 and 4-98: Matterport project uploaded, and Matterport Space ··························· 138
Figure 4-99: Matterport project space ······································································ 139
Figures 4-100 and 4-101: white circles and different views of a project ······························· 140
Figure 4-102: Doll house view ·············································································· 141
Figure 4-103: Floorplan view ················································································ 141
Figures 4-104 and 4-105: Measurements can be taken from any view mode ·························· 142
Figure 4-106: Matterport project details ···································································· 143
Figure 4-107: Matterport add on features ·································································· 144
Figure 4-108: Share and Invite page ········································································ 145
Figure 5-1: Master bedroom in Reunion House ··························································· 148
Figure 5-2: Reunion House master bedroom and six targets ············································· 152
Figures 5-3 and 5-4: Professional scanner scan and moving path ······································· 153
Figure 5-5 and 5-6: Smartphone scan moving path and scan result ····································· 154
Figures 5-7 and 5-8: Smartphone device tilted up and scan result ······································ 154
Figures 5-9 and 5-10: Smartphone device tilted down and scan result. ································· 155
Figure 5-11: Three smartphone scans merged with CloudCompare ···································· 155
Figure 5-12: Manual align smartphone scans (red) with professional scans (green). ················· 156
Figure 5-13: Section layout on point clouds ······························································· 156
Figure 5-14: Section 1 in AutoCAD ········································································ 157
Figure 5-15: Section 2 in AutoCAD ········································································ 157
Figure 5-16: Section 3 in AutoCAD ········································································ 158
Figure 5-17: Measurement example ········································································ 159
Figures 5-18 and 5-19: Blind area ··········································································· 160
Figure 5-20: Master bedroom floor to ceiling measurement example ·································· 161
Figure 5-21: Master bedroom height measurement ceiling tracing detail ······························ 161
Figure 5-22: Master bedroom height measurement floor tracing detail ································ 162
Figure 5-23: Tracing top and bottom points in section 2 with line command ·························· 163
Figure 5-24: Measuring point thickness with dimensional tool ·········································· 163
Figure 5-25: Layer of points from Leica RTC 360 ························································ 166
Figure 5-26: Ceiling detail in low point density scan section 2 ·········································· 171
Figure 5-27: High density iPhone scan (red) and RTC 360 scan (green) ceiling detail ··············· 172
Figure 5-28: Low density iPhone scan (red) and RTC 360 scan (green) closet detail ················ 173
Figures 5-29 and 5-30: Low density iPhone scan and RTC 360 scan section 1 and detail ··········· 173
Figure 5-31: Merge scans from smartphone aligned with professional scanner point clouds ········ 174
Figure 6-1: Chapter 6 overview diagram ··································································· 182
Figure 6-2: Smartphone scans manually aligned with professional scanners’ scan in AutoCAD. ·· 185
Figure 6-3: Low density iPhone scan (red) and RTC 360 scan (green) closet detail ·················· 186
Figure 6-4: High density iPhone scan (red) and RTC 360 scan (green) ceiling detail ················ 187
Figures 6-5: Low density iPhone scan (red) and RTC 360 scan (green) section 1, and detail ······· 187
Figure 6-6: Mirror reflection captured by smartphone ···················································· 189
Figure 6-7: Refracted landscape captured through awning windows ··································· 189
Figure 6-8: iPhone scanned front yard at Reunion House with a “spider leg” feature. ··············· 191
Figure 6-9: “Spider leg” feature detail in iPhone captured point cloud. ································ 192
Figure 6-10: “Spider leg” feature photograph. ····························································· 192
Figure 6-11: RTC 360 works outdoor. ······································································ 193
Figure 6-12: RTC 360 scanned outdoor environment at Reunion House ······························· 193
Figure 6-13: Cloud to cloud distance computation result in CloudCompare ··························· 195
Figure 6-14: Doll house view of iPhone scanned Reunion House. ······································ 196
`
x

Figure 6-15: Downloadable Reunion House videos and photos from Matterport webpage. ········· 197
Figure 6-16: Other deliverables from Matterport webpage. ·············································· 198
Figure 7-1: 360 degree photograph of Reunion House master bedroom. ······························· 203
Figure 7-2: 3D scan of Reunion House master bedroom. ················································ 203
Figure 7-3: Historic photograph of Reunion House ······················································· 204
Figure 3-1 Proposed methodology ·········································································· 206
Figure 7-4: Hoose Library of Philosophy at Mudd Hall, USC. ·········································· 207
Figure 7-5: Point cloud scan of Hoose Library of Philosophy. ·········································· 208
Figure 7-6: Merged scan of Hoose Library of Philosophy at Mudd Hall through CloudCompare. · 209
Figure 7-7: Cloud to cloud distance computation result in CloudCompare ···························· 210
Figure 7-5: Overlapping professional scanner point cloud with iPhone scanned point cloud. ······· 211
Figure 7-6: Qualification of smartphone scan for heritage conservation tasks. ························ 213
Figure 7-7: Game engine used in architecture ····························································· 215
Figure 7-8: Virtual reality in architecture ·································································· 215
Figure 7-9: Heritage conservation task diagram. ·························································· 217

`
xi

Abstract
3D scanning, as a digital documentation and analytical tool, has been practiced for decades to support the
decision-making process for heritage conservation. Site surveying, condition monitoring, documentation,
educational presentation, and other traditional aspects of heritage conservation are supported using 3D
scanning data today. Considerable previous literature has demonstrated many of its abilities to support
conservation goals and its great potential for expanding capabilities. However, the high cost of current
professional 3D scanning often becomes a deterring factor for less well-funded projects and projects with
accessible related issues.

The development of smartphones and tablets equipped with built-in sensors, such as cameras and LiDAR
systems, has opened up the possibility of using them as cost-effective tools for gathering geometric data for
cultural heritage. This alternative approach could revolutionize the way heritage conservation professionals
collect and visualize data. The enhanced accessibility of scanning with smartphones provides a chance for
students, professionals, and the public to engage in the data collection process, thus fostering the sharing of
knowledge all over the research value chain.

The ability of the proposed methodology with selected smartphone application and computer software to
fulfill heritage conservation goals was tested in test scans and the case study. The case study is the master
bedroom of Reunion House designed by Richard and Dion Neutra, which was built in 1951. Specifically,
evaluation of iPhone 13 Pro built-in LiDAR system accuracy through iOS application SiteScape, and the
digital products’ availability and effectiveness from such device was conducted by comparing to the
performance of a Leica RTC 360 professional scanner. Furthermore, smartphone competence in creating
360-degree photographic virtual tour was demonstrated with Matterport Capture application. Through an
analysis of the acquisition process, registration, and point cloud quality, the strength and limitations of the
smartphone scan method are discussed. The point cloud acquired with iPhone 13 Pro exhibited a 2% of
`
xii

error within the range of 17 feet compared to the point cloud captured by the professional scanner. The
iPhone 13 Pro acquisitions were shown to be an accessible solution to quickly acquire spatial information
with a lower level of detail with a low-cost.

Keywords
Smartphone, 3D scanning, Point clouds, Heritage conservation, LiDAR, Handheld mobile laser scanning
Hypothesis:
In the field of heritage conservation, adequate 3D spatial data for generating floor plans, building condition
recording, and digital reproduction for educational purposes can be acquired by using the capabilities of a
smartphone rather than using expensive professional 3D scanning equipment.

Objectives:
1) Identify the suitability of the new 3D smartphone-based scanning method for heritage conservation
purposes.
2) Develop a digital documentation workflow for heritage conservation that involves low-cost
scanning.
3) Test the suitability of several overlapping point clouds to create a more precise 3d model

1

Chapter 1 Introduction
“Historic preservation is an important way for us to transmit our understanding of the past to future
generations” (National Park Service, 2021). Various previous practices, from archival research to site
surveys, building maintenance to public education have been carried out to support the purpose of
heritage conservation. The growth of technology in the last few decades has created many alternative
methods for reaching a deeper understanding and recording of the existing built environment. 3D
scanning instruments and techniques have become available to improve the quality of data collection for
preservation-related works. However, not all cultural heritage projects can benefit from many of these
advanced technologies. The issue of authorized heritage discourse (AHD) lies in both intellectual and
physical accessible levels that constrains the debates about the meaning, nature, and value of heritage
(Smith, 2006). From an operational perspective, the typically high expense of professional geometrical
data capture and processing places obstacles for people of all levels to access and capture information. To
overcome concerns about the affordability, accessibility, and usefulness of 3D scans of heritage sites, a
method of using the built-in systems on a smartphone was tested at Richard and Dion Neutra’s Reunion
House. The methodology is expected to support local-level heritage conservation projects and heritages in
endangered conditions and inaccessible locations with low-cost and easy operation on obtaining valuable
data.
1.1 Heritage Conservation and Technology
Heritage conservation benefits from advances in technology in various aspects. Advanced tools support
people in the field to protect endangered valuable cultural heritages with better performance.
Conservation goals are achieved, and the results led to the next level with quickly updated technologies.
At the same time, many participants in heritage conservation may not benefit from the technology
because of limited funds. Cost-effective concerns constrain the scope and depth of heritage conservation

2

projects. Alternative options are available to acquire useful data with an affordable device and accessible
operation.
1.1.1 What is Heritage Conservation?
Heritage conservation (HC) is a discourse seeking to preserve, conserve and protect buildings, objects,
landscapes, or other artifacts of meaning and identity-making, socially and culturally (Duluth
Preservation Alliance, 2022). People look at history, ask questions, and learn new things about
themselves through the lens of heritage conservation. It is important for people to transmit their
understanding of the past to future generations.

The National Park Service has produced standards and guidelines that govern preservation efforts at the
national, tribal, state, and local levels to ensure uniform procedures. The Secretary of the Interior's
Standards for the Treatment of Historic Properties provides guidance for heritage conservation
professionals to conduct their work (Norman et al., 2018b). The guidelines address four overarching
treatments under preservation action: preservation, restoration, rehabilitation, and reconstruction with
standards. Guidelines provide general design and technical recommendations to assist in applying the
Standards to a specific property. Together, they provide a framework and guidance for decision-making
about work or changes to a historic properties (National Park Service).

1.1.2 What do we conserve?
Works of art and other elements of human creativity are preserved and protected through the recognition
of their cultural significance and the condition of their integrity. This recognition has shifted from
individual structures to entire territories, with cultural content seen as essential to their preservation
(Jokilehto, 2021). It is this recognition that has allowed for the preservation and protection of both
tangible and intangible heritage. Tangible heritage that carry the intangible heir from past generations

3

contribute to forming an identity within social and cultural life. UNESCO defines cultural heritage
broadly as “the legacy of physical artifacts and intangible attributes of a group or society that are inherited
from past generations, maintained in the present, and bestowed for the benefit of future generations”
(UNESCO). The value of heritage sites, places, and objects are required to be understood with the
context. Studying the physical heritage provides crucial evidence to reinforce the identity formed from it.
The conservation of substantial property demonstrates a recognition of the necessity of the past and its
value as societies of different times cherish values with various standards, the measure of heritage
changes. The physical attributes of heritage may also be altered in the future; thus, recording current
geometric conditions can be beneficial for future generation studies.
Heritage is not solely confined to material evidence, while the cultural significance being valued needs
the vessel of substantiality. The information in the conservation process is framed by a conservator’s
ability, which will be limited to the cognitive framework in which conservators operate (Sully, 2008).
Sully’s statement does not only apply to the post-colonial content of conservation but also a universal
account that value is subjective. What is valued in the current social context may change in the future;
material evidence helps lock the information from the past from being ignored or miss interpreted. Both
the materiality and cultural significance of heritages shall be conserved.
1.1.3 How do we conserve?
Preservation, restoration, rehabilitation, and reconstruction are four categories of treatments for Heritage
Conservation actions as defined by The Secretary of Interior Standards for the Treatment of Historic
Properties. The four treatments serve different goals, including maintaining and retaining existing
historical materials, adaptive reuse, returning to a particular period of significance, and recreating the
vanished property (National Park Service, 2022b).

Before any conservation work begins, a thorough investigation of the property is required. Archival
research and site surveys are conducted to expand the knowledge about the area. The investigation reveals

4

the cultural and historic background of the sites, as well as their current physical condition. Based on the
weights assigned to cultural and historic value, a period of significance and character-defining features
(CDF) are determined. To serve different goals of conserving a building, alteration, designation,
maintenance, education and other actions can be taken corresponding to the client’s request.

The National Park Service’s Preservation Brief 35 Understanding Old Buildings: The Process of
Architectural Investigation indicates that documentation of a building should be done before any other
preservation process. (Travis, 1994). A documentation that combines both graphic and written description
provides the opportunity of studying a structure without visiting it. The records can be used to find out
about things from the past that might be too far away, too hard to get to, or have already been lost.
Documentation is also a backup to present and to reconstruct for unforeseen damage to significant
structures. The Historic American Building Survey (HABS) was established by the National Park
Services, the Library of Congress, and the American Institute of Architects in 1933. HABS recognizes
that words alone are not enough to record and explain buildings, the pictorial representation is
indispensable in this process. As an archive program, HABS is required to ensure the clarity, reliability,
durability, and standardization of documentation. Quickly developed digital technologies change at a
rapid pace, often before data can be migrated or stored. Thus, digital technology is considered only
suitable as a tool to produce documentation, but not as a final product.

Documenting a property is one of the most intriguing aspects of preservation, and the study may be
compared to solving a riddle. The potential significance of a structure is estimated referencing the four
criteria listed in "National Register Criteria for Evaluation": association with events, significant person,
distinctive architectural style, and history or prehistory (Norman et al., 2018a). Integrity is one of the
qualities used to identify a property's historic importance. The National Register program developed
Location, Design, Setting, Materials, Workmanship, Feeling, and Association as seven aspects to be
evaluated in determining a structure’s integrity (Norman et al., 2018a). Thematic and historic context is

5

another important element in accessing historic significance, referring to the cultural setting in which the
property was formed, as well as its subsequent history. The thematic framework, which includes eight
categories that are built on people, time, and place, is encouraged by the National Park Services in the
evaluation process (Norman et al., 2018a). This information can be found through deep research from
photographs, old newspapers, legal documents from city maps and lithographs, Sanborn fire insurance
maps, oral traditions, and observing the building itself.
An historic context statement is the document that provides an overview of the influence of construction
traditions, development eras, and shareable character of places based on research and evaluation
standards. It is a required document for many nominations and designations at various levels (Norman et
al., 2018a). A nomination for the National Register of Historic Places will go through a review process
from the states historic preservation office (SHPO) and the Secretary of the Interior in Washington, D.C..
Once approved, the property will be listed in the National Register and published in the Federal Register
(Norman et al., 2018a). The National Register of Historic Places represents national recognition of a
historic property, but it does not protect the structure from alterations or demolition (Norman et al.,
2018a). Properties with exceptional value for the United States can be designated as National Historic
Landmarks with additional evaluation and documentation. Both nomination and designation of the
National Register and National Historic Landmarks require intensive investigation, documentation, and
description to the property (Norman et al., 2018a).

Before any conservation treatment is undertaken, a Historic Structure Report (HSR) should be created to
serve as a planning document. An HSR is a thorough record of existing historical research and resources
as well as existing conditions (Slaton, 2005). It provides a forum to identify historic fabric and the means
to minimize its loss, damage, or any adverse effects upon it (Norman et al., 2018a). From an
understanding of the historic fabric, long-term alternative actions and their impact on the site as a whole
can be explored in the planning phase (Burns, 2003). Its overall substance is a two-part narrative of the
structure's developing history and recommendations for its treatment and application, as well as

6

references to previous work. The essential is laying the groundwork for future practice to be done with
precision, integrity, and sensitivity to the structure's historic and cultural value (Norman et al., 2018a). An
HSR includes historic, architectural, engineering, analysis, landscape, archaeology and furnishing
sections. The current condition of a structure is the main concern for the architectural data section (Burns,
2003). Buildings deteriorate overtime, and accurate measured drawing should reflect those changes.
Recommended methods of recording a building’s current condition including measured drawing, large-
format photography, computer aided drawings, and videography (Arbogast, 2010).

All the efforts mentioned above dedicated to the conservation of Heritage resources contribute to the
eventual purpose of benefiting future generations (UNESCO). We acknowledge the value of both cultural
and historic significance together with the physical condition of heritage. The investigation and recording
of tangible material evidence serve the disclosing of information. Constant maintenance ensures this
material evidence deteriorates at a slower rate and education exposes such knowledge to a wider
audience.
1.1.4 Heritage Conservation and technology – technology for Heritage Conservation
When the Mount Vernon Ladies Association saved George Washington’s house in 1858, heritage
conservation deployed a very different workflow compared to today’s preservation (Mount Vernon
Ladies’ Association, 2022). Not only is the workflow growing more and more comprehensive, the
innovation of technology also drastically changes the way people learn about the built environment. From
measured drawing and photography to computer aided design (CAD) and 3D scanning, the advancement
in technology assists heritage conservation professionals to achieve accurate and effective results.
1.2 Digital documentation
From measured drawings and printed photographs to computational drafting and 3D scanning, digital
technologies are gradually becoming the dominant tools of documentation. Under the assistance of

7

traditional methods, digital technologies can record capture detailed information of historic buildings,
sites, and objects. The quick and accurate capture of exact measurements of a structure’s dimensions,
physical characteristics, and details of its construction provide a platform for heritage conservation
professionals to create detailed preservation plans (Gray, 2022).
1.2.1 2D documentation
Measured drawing and photography are the two most commonly used documentation methods. Paper
based drawing and large-format black and white photographs are especially praised for the permanence
and accessibility for long term archival storage (Library of Congress, 2011) (Figure 1-1). They are also
the preferred file format for HABS. Original drawings and negatives were scanned and digitized to
Library of Congress Digital Collections.

Figure 1-1 Library of Congress HABS Online Database Digital Collection (Library of Congress HABS
Online Database Digital Collection, 2022)

Measured drawings

8

Measured drawings are a detailed form of architectural and engineering documentation that accurately
portrays a three-dimensional structure or site in two dimensions (Fig. 1-2). This process involves
translating the cultural values of a three-dimensional object into two-dimensional illustrations and serves
many purposes, such as planning restoration or rehabilitation work, recording a structure facing imminent
demolition, aiding in the normal maintenance of a structure, protecting against catastrophic loss, or as part
of a scholarly study (Akboy-İlk, 2017). These drawings can be utilized for a variety of purposes, such as
planning for restoration or rehabilitation of a structure, recording a structure facing imminent demolition,
aiding in normal maintenance, or as part of a scholarly study (Norman et al., 2018a).

9

Figure 1-2: Measured drawing example

The production of a measured drawing involves making decisions about the significance of the structure
and the scale, features, and level of accuracy to be included. Documents, hand measurements, and
photographs are the main sources of information used to capture dimensions (Burns, 2003). Dimensions
recorded on field notes are the primary source, and they often contain more dimensions than are included
in the final drawing. This method helps architects and conservators become familiar with an object and
discover subtle aspects (Eppich & Chabbi, 2007). The smallest unit of measurement in a drawing is
determined by the scale. For example, the most common architectural scale is 1/4” = 1’-0” with a smallest
unit of 1”, while hardware, tools, and moldings can be measured down to 3/32” in a 3”=1’-0” scale
(Burns, 2003).

Measured drawings produced by hand are one of the costliest types of architectural and engineering
documentation due to their prolonged production time. When budget or time is limited, sketch plans can
be used in place of measured drawings. Although they may not be accurate in scale, sketch plans should
show elements in their correct proportions relative to one another (Burns, 2003; Norman et al., 2018a).
Film
Photography is the most often used means of documention. Photographs are simple to understand and can
convey information that other types of documentation cannot. It is capable of conveying three-
dimensional features, spatial linkages, present situations, texture, and context, which are difficult to
express in writing or painting (Burns, 2003). Careful photography can be both aesthetically pleasing and
informative (Burns, 2003). While it may not be a replacement for drawings, histories, or even viewing a
structure or site in person, it offers a unique perspective and a way to keep structures alive in the future.

10

Other than drawings and written descriptions, large-format photography is the official format for HABS
documentation of structures and buildings. HABS encourages large-format photograph not only because
it capture more information, but also because of the stability of black and white negatives. With archival
longevity as the goal, original large-format negatives will survive more than 100 years with careful
handling and storage, then produce prints with no degradation of the image (Burns, 2003). The National
Park Service also published Heritage Documentation Programs HABS/HAER/HALS Photography
Guidelines in 2015 to instruct conservators taking photography for documentation purposes, including
equipment, view, format, etc., regarding various types of built environments (Burns, 2003). Architectural
photography should follow the same shared principles. An understanding of the subject, proper lighting,
scaling tools, and aesthetics are common standards for photography taken by either in large-format,
35mm, or digital cameras.
The use of digital photography has become the accepted method for recording the current state of historic
sites. It can be used to supplement or replace hand-drawn sketches by being incorporated into computer
drawings. Another method of photography is rectified photography; as Getty’s RECORDIM: Guiding
Principles & Illustrated Examples defined “[it] is the process of photographing a facade by aligning the
images to be as parallel as possible to the section of facade to be recorded (Eppich & Chabbi, 2007).
Using this method, it is possible to obtain the dimensions of a building from a photograph rather than
having to take time consuming measurements on the site. X-ray photography and radar on building
structure can identify materials and structures behind the surface (Norman et al., 2018b).
Two-dimensional photography is the dominant method of digital documentation for its easy process to
capture, editing, share, and view. In building survey phone apps such as Fulcrum, a photograph of the
property is required at the end of each description. However, a photograph can only capture information
from one point of view. More details of the building need to be pieced together with photographs from
various perspectives and distances. Even with such detailed documented photographs, one still needs to
transform the two-dimensional visual information, through their mind, into three dimensional objects.

11

Therefore, directly documenting properties with three-dimensional spatial data reduces the deviation
created by differences in human perception, which optimizes the understanding of architectural features
and their significance.

Figure 1-3: Film photography of Reunion House living room (USC Library, 1951)
1.2.2 360-degree photographs
360-degree photograph is a controllable panoramic photo taken on the original point from which the shot
was taken (Panoraven, 2021; TechTarget Contributor, 2016). Shooting photographs at one location of
many angles, a full spherical view was then created in a raw 360-degree photograph (Fig. 1-4) (Burns,
2003; Panoraven, 2021). With software or application, the 360-degree photograph can be navigated in
different directions, as one standing and looking into different directions (TechTarget Contributor, 2016).
360-degree photographs are different from 3D stereo photographs (Fig. 1-5), which adds a third
dimension to photographs. To view the depth inside a 3D photograph, special 3D glasses are used.

12

Figure 1-4: Raw 360-degree photograph (Panoraven, 2021)

Figure 1-5: 3D photograph (Bak, 2017)

13

360-degree photographs can be taken by a smartphone with a certain application, a 360 camera, or a
Digital Single Lens Reflex camera (DSLR) (Panoraven, 2021). 360 cameras came from different
companies and make, for example, Insta 360, Ricoh, and Matterport. Smartphone applications that
capture and generate 360-degree cameras including 360 Pro, Panorama 360, Matterport Capture.

360-degree photographs are common in the survey and real estate industry. The most well-known
examples are Google street view and real estate virtual tours (Fig. 1-6). Even though 360-degree
photographs provide more spatial experience, it does not obtain any depth data.

Figure 1-6: Google street view of Forbidden City

Matterport 360-degree photographs are also commonly used in generating virtual tours. As shown in Fig
1-7, through a browser, one is able to move between different camera capture location and visit 360-

14

degree view of the place described. Assisted with Virtual Reality equipment, 360-degree photographs can
be experienced in virtual environment.

Figure 1-7: Michael White Adobe virtual tour (AQYER, 2022)
1.2.2 3D documentation using scanners
The use of 3D scanning is becoming increasingly popular in heritage conservation as it improves
efficiency and accuracy in acquiring, analyzing, and presenting information. 3D scanners nowadays can
be equipped with a mobile device or a drone to complete tasks within several hours that used to take
weeks by labor-intensive measurements. Through algorithms and mechanisms, precise data with high
accuracy and resolution is captured and presented. However, the high expense of scanning and experts
operating that data place barriers for projects with fundraising issues.

The two barriers are clear: the high expense of renting or owning high-end scanning equipment and the
sophisticated operation and registration process of visualizing point clouds data. The two challenges are
so interrelated that they usually come simultaneously. The issues are expected to be mitigated with a

15

smartphone equipped with a relatively high-resolution camera, embedded LiDAR system, and installed
data processing applications. LiDAR began to be installed in smartphones in 2020 with Apple company
releasing its new model of iPhone 12 Pro (Luetzenburg et al., 2021). Appropriate use of a smartphone can
assist property owners, cultural resource managers, and other stakeholders with the initial survey,
documentation, maintenance, and education to evaluate the necessity of hiring experts on any of these
areas with the higher expense and explore the possibility of achieving these tasks with an acceptable
amount of detail in an affordable and accessible method.

Another scenario that can benefit from the methodology is for heritage sites endangered or located at an
inaccessible place. Heritage in a war environment or with difficulty in transportation is greatly
endangered. These resources beg for more attention, but due to their inaccessibility, it is hard for heritage
conservation professionals to protect the site and pass the heritage onto future generations. Smartphone
scans allow non-technicians to acquire geometrical data of such heritage property and share point clouds
with specialists remotely.

Acquiring three-dimensional spatial data from architectural features can be achieved through 3D scanning
techniques. While with the development of science and its application, there are dozens of 3D scanning
equipment and techniques available for different purposes. There are a number of 3D scanning methods
based on different working principles, in various working environments, and multiple levels of precision
and accuracy. The most well-developed and extensively deployed are 3D scanners working in
triangulation, structured from light, photogrammetry, pulse, Phase-Comparison, and 3D photography
equipment such as Matterport and 360 cameras. For 3D documentation methods, point clouds are a
universal file format that further data processing can be based (El-Ashmawy & Shaker, 2014).

Point clouds as its name indicated, are clusters of data points. This group of points with each point
defined in Cartesian coordinates (X, Y, Z) describes a three-dimensional shape. Point clouds are different

16

from a surfaced 3D model of Building Information Modeling (BIM) because it does not have a surface
and does not include information beyond spatial data (Park & Lee, 2019). With software such as Recap
Pro, a point cloud can be used to generate a surfaced mesh model. The measurements using mesh models
are believed about 2%–3% smaller than those using direct point clouds (Fig. 1-8) (Park & Lee, 2019).
The triangular mesh quality can vary significantly in terms of the point density, the algorithm, or the
complexity and shape of the object surface (Park & Lee, 2019).

Figure 1-8: Point Clouds, Control Mesh, Result visualization of a rabbit (Yoon, 2006)

A point cloud is a series of 3D points. However, most 3D modeling software programs are designed to
handle meshes, and while going from a point cloud to a mesh is easy for a simple object, it is extremely
difficult separating it into multiple objects (Fig. 1-9.) Point clouds can also be associated with textures
(Fig. 1-10)

17

Figure 1-9: Conegliano Italian Synagogue - Point Cloud (Caine, 2019)

18

Figure 1-10: Same point cloud data with different processing in reconstructing surface (Giraudot, 2022)

Triangulation Scanner
The triangulation scanner is named because of its working principle that the emitted laser and the
reflected laser light form a triangle (Acuity Laser, 2022). Through calculating the triangulation of the
position of a spot or stripe of laser light, the scanned object can be calculated to shape. More accurately
speaking, the light of a laser through a rotating mirror shoots onto the subject. The mirror turns to deflect

19

the light to thoroughly scan around the subject. In this process, each reflected beam of laser is focused
onto the sensor or camera by another lens (Fig. 1-11). The location of the point on the sensor, the known
separation (D) between the lens and the mirror, and the recorded angle of the mirror combined provide a
3D coordinate based on basic trigonometry (Boardman & Bryan, 2018).

Figure 1-11: Working principle of triangulation scanners (Historic England, 2018)

Structured Light scanner
Structured light scanners, illustrated scanners, and some handheld scanners all share similar working
principles as the triangulation scanners with some variations. The difference between structured light
scanners with triangulation scanners is the amount of light or laser shot onto the subject. A structured
light scanner emits a “sequence of organized patterns of light,” projecting on the object’s surface (Fig. 1-
12) (Wachowiak & Karas, 2009). The distortion of the light pattern is analyzed and the distance of every
point is calculated using the surface topography (Raychev et al., 2017)

20

Figure 1-12: Working principle of Structured Light scanners (Bitfab, 2022)

Structured Light scanners are most often used in common range heritage documentation and both types of
scanners capture excellent surface and color data (Boardman & Bryan, 2018; Wachowiak & Karas, 2009).
Its relatively low cost compared to other professional scanners ($100,000–$200,000), portable system,
together with the accurate spatial registration make these types of scanners highly desirable for heritage
documentation work (Wachowiak & Karas, 2009).

While its scan quality is largely determined by the control of environmental light, which adds difficulty to
operation (Boardman & Bryan, 2018). It exhibits the best capability in darkened situations where the
emitted and any ambient light are especially evident. The limitation of these scanners shows up when
lighting conditions are not preferred. These conditions include an unclear view from both lenses to
objects, deep undercuts on the object surface (where light cannot reach), highly reflective surface,
reflectance, and transparency surfaces property, and ambient illumination (Agnello et al., 2005;
Boardman & Bryan, 2018; Wachowiak & Karas, 2009). In addition, the triangulation and structured from
light scanner work with a small to medium range (<10m), and are not suitable for large-scale architectural
structures and topographical surveys (Wachowiak & Karas, 2009).

Photogrammetry

21

Digital photogrammetry was first proposed by Ian Dowman in 1984 to map the topography of terrain
using satellite imagery. Three-dimensional information is calculated and measured from two-dimensional
photographs. Calculation of triangulation is the working principle of digital photogrammetry.
Photographs taken from different locations have different “lines of sight” between each camera site to the
object (Douglass et al., 2015). Through a mathematical process of the angle, location, length, and distance
information of the line of sight, three-dimensional data can be produced. The quality of photogrammetry
largely depends on the photographs used for calculation, including the photographs’ resolution and the
area of overlapping on each abutting photograph (Fig. 1-13). Photogrammetry software for smartphones
was also developed in the past decade. People are able to capture 3D data from a portable and affordable
device.

Figure 1-13: Working principle of photogrammetry (Shet, 2022)

Pulse (TOF)

22

According to Historic England’s 3D Laser Scanning for Heritage Advice and Guidance on the Use of
Laser Scanning in Archaeology and Architecture “Pulse scanners use what can be considered to be the
most straightforward technology: a pulse of laser light is emitted and the time it takes for the return flight
is measured” (Boardman & Bryan, 2018). Light detection and ranging (LiDAR) is considered a typical
pulse scanning method.

This functionality is achieved through a sophisticated mechanism timing the receiving of the laser light
and a precise mirror on a rotation system. The system can be rotated 360° around a vertical axis and
between 270° and 300° around a horizontal axis (Boardman & Bryan, 2018). Forming almost a complete
sphere of view, the rotation system provides a great advantage to pulse and Phase-Comparison laser
scanners compared to triangulation and structured light scanners (Boardman & Bryan, 2018) (Fig. 1-14)

Figure 1-14: Principles of laser scanner data acquisition, showing the example of TLS (Jaboyedoff et al.,
2012)

23

Similar to triangulation and structured light scanners, scanners with a pulse working principle can neither
work on translucent nor reflective surfaces. The process of receiving signals back from such a surface
may produce degradation of the quality of the range data, thus generating two critical issues for the
geometry evaluation: a bias in the distance measurement, as well as an increase of the noise level
(Agnello et al., 2005; Andrews et al., 2003; Haddad, 2011).

In the past, pulse scanners were criticized for their slow scanning process, and the last pulse of the laser
cannot be emitted until the earlier one has been received. In 2015, Leica company released ultra-high-
speed scanners with rates of 1MHz (1 million points per second) (Leica Geosystems AG, 2022c); later the
same year, RIEGL released their VZ-400i Terrestrial Laser Scanner, achieving 1.2MHz pulse repetition
rate (RIEGL, 2022), mainly shortening the time per scan.

A LiDAR scanner is commonly equipped with a tripod, mobile device, or a drone because the energy
emitted in a single pulse of laser light is strong enough to support the system scanning from a great
distance, typically up to 1km but in some cases up to 6km (Riegl VZ-6000). The vital energy in one laser
pulse enables it to pass through a tree canopy and reach the terrain in airborne scanning. This
characteristic also gives it an advantage over Phase-Comparison scanners in bright daylight (Boardman &
Bryan, 2018).

Starting in 2020 when Apple company released their iPhone 12 Pro that is embedded with a LiDAR
system, a growing number of iOS applications such as SiteScape enable geometrical data to acquire
functions. Users can obtain 3D data and upload point clouds to a cloud based platform for further editing.
SiteScape app allows either free or paid user licenses, which leads to different levels of services. With a
free user account, one can scan with controllable parameters, and upload one data set at a time to the
Cloud (Putch, 2022). While for a paid version, users can easily merge several scans together to generate a
larger area of space, as well as multiple data being uploaded to the Cloud.

24

Phase-Comparison
Phase-Comparison scanners are similar to pulse scanners in that they are based on the round trip of the
laser pulse. The difference is that instead of timing the roundtrip of a single pulse of laser light, Phase-
Comparison scanners measure the wavelength difference between the laser emitted and the laser reflected
(Daneshmand et al., 2018). Instead of a single laser pulse, Phase scanners emit a constant laser beam to
the scanned surface. As the laser light shoots on the surface, some portion gets absorbed while others
reflect to the scanner with a changed wavelength and frequency. The shape was therefore calculated
through the difference in frequency (Suchocki et al., 2021). In HC projects, Phase-Comparison scanners
performed well in capturing damage under surfaces. It can collect data much faster than structured from
light and pulse scanners, but because the energy is lower and frequency can be disturbed, their effective
distance is shorter. Due to the working principle of measuring frequency difference, phase-based scanners
can be affected and create more "noise" and inaccurate data (Existing Condition, 2022)

1.3 Accuracy versus precision
Scanned data is commonly evaluated based on their ability to achieve certain goals. To better understand,
the scanned data’s quality, the metrics such as accuracy and precision are proposed.

Even though commonly obfuscated, accuracy and precision describe different aspects of data. Accuracy
describes how close a measurement is to the true or accepted value. Precision is about how close
measurements of the same item are to each other (Fig. 1-15) (Exploring Our Fluid Earth, 2022). In the
case of scanning, accuracy can be determined as metric varied between scanners. On a product page,
companies would exhibit system accuracy. Taking Leica P30/40 as an example, the company stated the
scanner can capture points at a 3mm (0.12 inch) accuracy in a 50-meter (164 foot) distance (Leica
Geosystems AG, 2022b). While even conducting the same scan twice without changing the scan location,

25

the points obtained through scanners can be different. Points may not stand on the exact same spot as the
previous scan. However, the repeatability of points is not heavily weighted in achieving HC tasks. In
monitoring a structure or detecting cracks, the surface or line generated from the cluster of points is the
evaluation standard for precision. Even if points are not standing at the exact same location, the surface or
trend they generated is repeatable through various scans. Then it is said to be precise.

Figure 1-15: Precision VS Accuracy (St. Olaf College, 2022)
1.3.2 Accuracy of scanners
Scanners from different companies adopt different accuracy parameters. Scanners researched deploy an
accuracy range from 3mm to 15 mm. The accuracy of scanners is determined by working principle,
scanning range, and purpose. Accuracy data can be found in product page from company’s webpage
(Chart 1-1). A 3D accuracy chart extracted from product pages of 3D scanners. There is no clear
requirement on the accuracy for achieving specific conservation goals (Gray, 2022). However, if the

26

scanned data falls in low accuracy, it may not be able to review the movement of structure over time, or
to detect and exhibit cracks.

Table 1-1: 3D accuracy of 3D scanners

3D accuracy
ScanStation P50 4.4mm@50m
6.8mm@100m
ScanStation P40 3.2 mm @50
5.9mm @100m
ScanStation P30 3.2 mm @50
5.9mm @100m
RTC360 6.4mm @50m
12.5mm @ 100m
RTC360 LT 6.4mm @50m
12.5mm @ 100m
BLK360 6 mm@ 10m
8mm@20m
Focus Laser Scanner
350 Plus
2 @10m
3.5 @25m
Focus Laser Scanner
150 Plus
2 @10m
3.5 @25m
Focus Laser Scanner
350
2 @10m
3.5 @25m
Focus Laser Scanner
150
2 @10m
3.5 @25m
Focus Laser Scanner
S70
2 @10m
3.5 @25m
FARO Freestyle 2 ≤0.5 mm
0.5 mm at 1 m distance 5 mm at 5 m
distance 15 mm at 10 m distance
Riegl VZ 400i
3mm@50m, 5mm@100m
Riegl VZ 600i
3mm@50m, 5mm@100m
Riegl VZ 2000i
3mm@50m, 5mm@100m
Riegl VZ 4000i
15mm
Riegl VZ 6000i
15mm

27

1.3.3 Point Cloud data processing software
Point cloud data cannot be visualized, processed or edited without software. There are a number of
software programs that are designed to work with point cloud files. Some were written by scanner
companies, while others were developed by technicians. Among these software programs, most requires
paid user licenses, CloudCompare and Meshlab are two that offer free download and functions. Meshlab
emphasizes processing and editing of 3D triangular meshes; CloudCompare has more functions for 3D
point clouds initial data processing including clearing out noise, and merging and aligning multiple scans
in test scans and the case study (Table 1-2)

Table 1-2: Point clouds processing software information
Company Software Price Function
CloudCompare CloudCompare Free
CloudCompare is a 3D point cloud (and triangular mesh)
processing software. It can be used to compare between
two dense 3D points clouds or between a point cloud and
a triangular mesh (CloudCompare, 2022).
Autodesk
ReCap Free
Recap is an Autodesk spftware that helps designers and
engineers capture high quality, detailed models of the
real-world object (Autodesk, 2022).
ReCap Pro Paid
Meshlab MeshLab Free
MeshLab is an open-source system for processing and
editing 3D triangular meshes. It provides a set of tools for
editing, cleaning, inspecting, rendering, texturing and
converting mesh data, and making models for 3d printing
(MeshLab, 2022).
Leica
Cyclone Paid
Cyclone is a Leica Geosystem software that processes,
models and manages 3D point clouds (Leica Geosystems
AG, 2022a).
Cyclone Cloud
Cyclone Cloud is a centralized, cloud-based version of
Cylone (Leica Geosystems AG, 2022a).
CloudWorx
CloudWorx is a digital reality plugins for AutoCAD
systems (Leica Geosystems AG, 2022a).
TruView
Share and view point cloud data freely via the web or
desktop application (Leica Geosystem, 2022)
Map360
Map360 is a software suite for building forensic
investigation (Leica Geosystems AG, 2022a).
Faro
Scene Paid
Faro Scene focuses on 3D point cloud capturing, data
processing and registration (Faro, 2022b).
As Built Paid
Faro As Built is for CAD & BIM modeling and drawing
(Faro, 2022b).

28

BuildIT
Construction
Paid
Faro BuildIT Construction is a complete design software
solution for continuous construction verification (Faro,
2022b).
BuildIT Metrology Paid
Faro BuildIT Metrology is a quality control, tool building,
guided assembly and machine alignment (Faro, 2022b).
BuildIT Projector Paid
BuildIT Projector planning, generating and operating laser
projection projects (Faro, 2022b).
Visual Inspect Paid
Faro Visual Inspection helps access CAD data on a
mobile device for fast visual inspection (Faro, 2022b)
Zoller +
Fröhlich
LaserControl®
Scout & Office
Paid
Z+F LaserControl provides a range of filters,
measurement and registration tools that enables a high
differentiate processing of scan data and are the key to
filter, register and color 3D point clouds (Zoller +
Fröhlich, 2022).
SynCaT® - Mobile
Mapping Software
Paid
Z+F SynCaT is a Mobile Mapping Software (Trimble,
2022; Zoller + Fröhlich, 2022)
Trimble RealWorks Paid
Trimble RealWorks are automated tools and point cloud
specific workflows that allow users to import point cloud
data from virtually any source, then quickly process,
analyze and create high quality customer deliverables
(Trimble, 2022).
1.4 Richard Neutra and Reunion House as a case study
As one of the greatest architects of the 20
th
century, Richard Neutra is famous for his international style
practice in the United States. The Reunion House at Silverlake, Los Angeles, built in his later years,
combines Neutra’s signature architectural elements and was designed programmatically as a house with
separate quarters for grandparents and grandchildred and a central meeting space (Lamprecht, 2021). The
property is owned by the Neutra Institute for Survival Through Design and was designated as a City of
Los Angeles Historic-Cultural Monument in 2021.
1.4.1 A Short biography of Richard and Dion Neutra
Richard Joseph Neutra (April 8, 1892 – April 16, 1970) was born in Austria and moved to the United
States in 1923 at the age of 31. He grounded his life and career in Southern California and became a
prominent modernist architect famous for suburban single-family houses. Studying architecture in Europe
and working for Frank Lloyd Wright, Neutra developed his strong personal style by combining

29

international style and the United States situation. For example, instead of using expensive materials,
Neutra used wood structure and silver paint to mimic the appearance of metal (Los Angeles Department
of City Planning, 2008).

Dion Neutra (October 8, 1926 – November 24, 2019) was Richard Neutra’s son, a Modernist /
International style American architect and consultant based in Southern California. Growing up in an
architect family, Dion Neutra received training from his father at age 11. He then studied and graduated
from the USC School of Architecture in 1950 (Lamprecht, 2021; Los Angeles Department of City
Planning, 2008). After graduation, he worked in Richard Neutra’s firm until 1965, when he became a
partner. Following his father’s death in the 1970s, Dion took over the leadership of the firm. On
November 24, 2019, Dion Neutra died at his home on Neutra Place ("Reunion House") in the Silverlake
neighborhood of Los Angeles at the age of 93 (Lamprecht, 2021; Los Angeles Department of City
Planning, 2008).
1.4.2 Reunion House
Reunion House is a hillside residence built in 1950 and located at 2440 Neutra Place in Silver Lake,
California, near Los Angeles. Richard Neutra designed the house to accommodate grandparents and
visiting family members, hence the name Reunion House (Fig. 1-16) (Bahadursingh, 2021). The Neutra
family bought the property in 1963. Three years later it was transferred to Richard’s son Dion Neutra.
(Bahadursingh, 2021). The house consists of a master bedroom, a living room open to the front garden, a
guest bedroom for grandchildren, a study, two bathrooms, and a kitchen. The home is sited on a hillside
and now hidden by abundant vegetation. An offset stair led visitors from the street to the front door.
Alterations have been made by Richard and Dion Neutra based on needs over decades (Lamprecht,
2021). In 2021 Reunion House was added to Los Angeles City’s list of Historic Cultural Monuments.

30

Figure 1-16: Reunion house, Silver Lake, Los Angeles, Calif., 1951 (Block, 1951)
The house is nominated as "an excellent example of a single-family dwelling in the mid-Century Modern
architectural style, and a highly intact work by architects Richard and Dion Neutra" (Bahadursingh,
2021). Later the same year, a condition report and recommendation for the house was completed by a
group of students from the University of Southern California.
1.5 Conclusion
Heritage conservation aims at preserving tangible and intangible cultural heritages. The discourse
constructs two sets of heritage practices, one focuses on the management and conservation of heritage
sites, places, and objects, and the other is related to the visitation of sites and institutions within tourism
and leisure activities (Smith, 2006). In this process, investigations of both cultural and historic aspects
and physical conditions are carried out. Surveys, documentation, preservation planning, preservation
treatment, post-treatment documentation, maintenance monitoring, and public education in the
investigation of the physical condition of the sites are considered conservation tasks. Among these tasks,

31

scanning technologies can be appropriately used to assist with surveys, documentation, monitoring, and
public education purposes.

Figure 1-17: Heritage conservation task diagram

Scanning techniques were categorized types with visualization format and data acquiring procedures: 2D
documentation including measured drawing and film, 3D photographs, and 3D scanning using different
methods. Photographs with film medium or panoramic photos are typical 2D scanning, with a
subcategorization of cylindrical and spherical panoramic photos. 3D spatial data can be captured through
photogrammetry, structured light scan, triangulation scan, pulse scan, and Phase-Comparison scan, as
well as Matterport and 360-degree cameras. Each of them has its strength and weakness, while a
thoughtful combination of techniques can maximize the benefit and achieve a certain goal.

32

High accuracy, high resolution, and long range make professional 3D scanners favored by heritage
conservation academia. However, considering the high cost of scanning by technicians and equipment,
some organizations, individuals, and projects may not be able to afford when fundraising has long been
an issue. Sophisticated data acquiring operations also place barriers for endangered heritage at
inaccessible locations due to war or pandemic. Therefore, a hands-on and cost-effective scan method is
needed. As smartphones began to be embedded with high resolution digital cameras and LiDAR sensors,
spatial data acquired through 2D and 3D photographs, together with photogrammetry and LiDAR
scanning can be provided by a portable and affordable smartphone or tablet (Figure 1-18). This alternative
is waiting to be tested and workflow to be developed.

33

Figure 1-18: Smartphone Scanning Methods

34

Chapter 2. Literature Review
Advancement in science and technology has allowed scanning to be used in various industries; including
but not limited to industrial manufacturing, medication, and large-scale ecosystem monitoring. In the field
of Heritage Conservation, scanning is also widely used. 3D scanners work on different principles in
acquiring 3D point data. Major types of scanners are triangulation, structured light, photogrammetry, time
of flight (ToF), Phase-Comparison, and 360-degree photograph. Professional scanners’ capability has
been well evaluated with a significant number of past experiments and studies. Portable devices such as
smartphones had only been studied in limited works of literature. The potential of using smartphone
LiDAR sensors acquiring useful data shall be discussed. Available research, archive, city documents and
a report on Reunion House are introduced at the end.

This chapter will describe 2.1 3D scanning’s application in general fields of industry, 2.2 3D scanning’s
application in the field of Heritage Conservation, categorized with different working principles, 2.3 past
studies using a smartphone for scanning, 2.4 past research and documentation done for Reunion House,
and 2.5 summary.
2.1 Scanning for non-Heritage Conservation purposes
Scanning techniques perform their functionality in various industries. From object to terrain, 3D scanners
work different principles such as triangulation, LiDAR, Phase-Comparison exhibit their limitation and
strengths in field including industrial manufacturing, medication, ecosystem monitoring and etc.
2.1.1 Triangulation scanners application

35

Triangulation scanners work on the Law of Sin, which utilizes trigonometric triangulation of the angle
and distance of light or laser to determine the location of a spot in three-dimensional space.
Photogrammetry and structured light (Fig 2-1) are two most commonly used triangulation scanners; they
have been applied in various scientific fields, such as industrial manufacturing (Ikkala et al., 2022;
Karganroudi et al., 2023; Siwiec & Lenda, 2022), ecosystem monitoring (Hirtle et al., 2022; Mineo et al.,
2022; Nasiri et al., 2022; Rodríguez et al., 2022; She et al., 2022; Strunk et al., 2022; Ternon et al., 2022;
Wang, 2022; Whitehead et al., 2022; Zhang et al., 2022), urban planning improvements (Taniguchi et al.,
2022), medication (Douglass, 2022; Lauria et al., 2022; Shao et al., 2019; Shao et al., 2022), and
anthropology studies (Garashchenko et al., 2022). Photogrammetry is a scanning method relying on
software calculation with 2D images; Structure Light scanning is an on-site scanning approach with
projected patterned light onto surfaces. Both photogrammetry and structured light scanning work
on trigonometric triangulation.

Figure 2-1: Structured Light Scanner (Lievendag, 2017)

36

There are reports on applying photogrammetry in the modeling of forest canopy cover as various
ecological parameters of forest ecosystems (Zhang et al., 2022). The authors found that the integration of
photogrammetry, Sentinel-2 data, and ML models can optimize the generation of landscape-level scale
maps in a precise and fast fashion (Zhang et al., 2022). Ternon et al. focused on the possibility of using
photogrammetry to map the rocky reef under turbid environments. Combined with RGB, DSM, and
several spatial benthic terrain variables, the methodology of mapping through triangulated data provides
new perspectives to understand the relationships between the reef rock and benthic organisms(Ternon et
al., 2022). Both works of literature proved that photogrammetry can generate three-dimensional
geometrical data at a topography scale, even in unclear situations.

Figure 2-2: The three major processing steps of PCs. (a) The vegetation was extracted by using SVIs. (b)
Grass was separated from the canopy using CSF. (c) The separated canopy PCs were triangulated using
Delaunay triangulation. (Zhang et al., 2022)

Not only used for ecosystem analysis, triangulation scanning is also used for improving urban living
environments. Triangulation scanning was used by a group of Japanese scholars to digitally reconstruct
sidewalks. Subtle undulations and elevations can be detected from the digital twin generated from the
collected photographs, thus improving disadvantaged groups’ urban living experience (Taniguchi et al.,
2022). In industrial applications, triangulated data improves molding manufacturing workflow, reducing
deviation and optimizing innovative maintenance systems (Karganroudi et al., 2023; Vizzo et al., 2022).

37

In museum settings, triangulated surface models by Structured Light scanning are used to digitally
construct a fossil skull in order to compare the difference between the object, point cloud scanned model,
and the CAD-built model (Garashchenko et al., 2022). Photogrammetry and Structured Light scanners
were compared to computer tomography (CT) scans in medical science. The research revealed that both
techniques cannot provide internal structure information; the requirement of multiple scans or
perspectives is also time-consuming. The literature noted that both optical scan methods performed well
within great detail. It also mentioned the trend of using smartphone light detecting and ranging (LiDAR)
for cost-effective and daily operation purposes (Douglass, 2022).
As these works of literature exhibited, photogrammetry is able to provide an overview of large research
areas and small object details. Equipped with mobile and aerial devices, photogrammetry can also be used
in inaccessible locations. While under conditions such as wind, clouds, or hazy weather aerial
photography can be affected in process and quality. With a canopy or partially covered terrain or surface,
photogrammetry is unable to generate data beneath the canopy. Structure from light scanners, which also
work on triangulation calculation, has a better performance in a well-controlled dark indoor environment
than in bright exterior space.
2.1.2 LiDAR scanners application
LiDAR is the abbreviation of “laser imaging, detection, and ranging” (Taylor, 2019). With a beam of
laser travel from scanner to object and comeback, the distance can be measured with time taken in the
roundtrip.

38

Figure 2-2 : Faro Focus Laser Scanner (Faro, 2022)

LiDAR takes the advantage of the strong energy emitted with every single beam of laser, allowing
scanning over a long distance. Equipped with a tripod or a drone, LiDAR scanners are commonly
categorized into Terrestrial Laser Scanning (TSL) and UVA. Long-range TSL scanning can reach up to 6
kilometers range. In 2019, Riegl VZ-6000 scanner was used to measure the annual mass balance of a
Glacier (Xu et al., 2019). Large outdoor environments (Vizzo et al., 2022), forests (Sofia et al., 2022), and
cities (Setyawan et al., 2022) are scanned with a pulse LiDAR system for management, visualizing, and
modeling purposes. The strong energy embedded in the laser beam is able to reach the ground through the
vegetation canopy. Even though a filtering process is required, LiDAR scanners enable the possibility of
scan the ground surface without bushes, hedges, and trees, placing the technology at advantage comparing
to triangulation scanners (Gitbook, 2022).

39

Figure 2-3: LiDAR drone concept. (www.microdrones.com)

Beyond scanning from great distances, LiDAR is also used in industrial diagnosis (Gargoum et al., 2022;
Jovančević et al., 2017; Shu et al., 2022). Researchers used LiDAR system to analyze collision data for
roadside safety assessment purposes. LiDAR laser system was implemented in welding equipment in
improving manufacturing performance with its high accuracy and precision (Shu et al., 2022). Sharing the
same purpose of high-accuracy measurement, airplane exterior defects detection also made use of LiDAR
systems (Jovančević et al., 2017).

LiDAR scan technologies utilize lasers, which are commonly used for tide analysis or shallow clear water
submarine environment reconstruction, due to water absorption (Filisetti et al., 2018; Zhou et al., 2021).
Key factors impacting LiDAR signal in the marine environment were examined; the range of laser
transmission was determined as over 35 meters in clean seawater, The transmission distance less than 20

40

m in coastal seawater and the transport distance in turbid port water was approximately 5 m (Filisetti et
al., 2018).

Since 2020 when Apple released their iPhone 12 Pro and iPad Pro, which are equipped with LiDAR, a
rising trend in academia began to test its capability of scanning. Studies have been performed in the area
of geoscience (Bharadwaj et al., 2022; Tavani et al., 2022), transportation monitoring (Wang, 2022), and
ecosystem (Holcomb, 2021). The user-friendly communication design and rapid scan and processing
speed were acknowledged by researchers, proving the possibility of replacing professional scanners under
certain conditions (Holcomb, 2021; Wang, 2022). To evaluate its capability, the quality of captured data
from smartphones was assessed and compared with established ground-based 3D scans (Spreafico et al.,
2021).

Figure 2-4: iPad mounted on tripod, testing static configuration (Spreafico et al., 2021)

41

2.1.3 Phase-Comparison scanners application
A Phase-Comparison scanner projects constant waves of infrared light of varying length, by receiving
reflected waves from object surface; the difference between wavelengths is used to generate shape
information (Fig. 2-3) (Daneshmand et al., 2018).

Figure 2-5: Phase Comparison Scanner (Faro, 2022a)

Several applications of Phase-Comparison scanners have been proposed in ecological monitoring, some
focusing on tree canopy (Stanley, 2013), and others on tree metric measurement (Pueschel, 2013). It was
reported in the literature that the maximizing of sampling efficiency can be achieved with low scanning
time. However, high accuracy can result in the requirement of merging multiple scans to achieve a certain

42

volume (Pueschel, 2013). Phase-Comparison scanners are also used in crime analysis (Esaias et al.,
2020). A Faro scanner was used in comparison with the manual method of estimating bloodstain origin. It
validated the practical benefits of 3D scanning and reliability in BPA reconstruction documentation for
technical review and court presentation (Esaias et al., 2020).

Figure 2-6: Phase-comparison scanner scanned image (Pueschel, 2013).
2.2 Scanning heritage
Scanning heritage not only provides documentation for the site or property, it is also the base of the entire
planning work. Panoramic photos, triangulation scans, photogrammetry, structured light scan, LiDAR
scans, and Phase-Comparison scans can all be combined and applied to benefit the field of heritage
conservation.

43

2.2.1 Panoramic photo application in Heritage Conservation
Panoramic photo is a quick and low-cost acquisition method that stitches photographs to a view up to 360
degrees with no distortion or aberration (Shum & Szeliski, 1999). With computer software and distance
meters, a panoramic photo is valued with geometric information (d'Annibale et al., 2013). The medium
has been implemented in the workflow of creating virtual architecture (d'Annibale et al., 2013).
Panoramic photographs were practiced with structure from motion and multi-image spherical
photogrammetry techniques in producing virtual reality tools and proved their ability in reconstructing
virtual scenarios (d'Annibale et al., 2013).

Figure 2-7 Panoramic photographs of the Colosseum in Rome (d'Annibale et al., 2013).

44

Similar research had also been done in revealing the promising ability of panoramic photographs when
working with spatial data obtained from TSL scanners (Salemi et al., 2005). Several historic buildings in
Europe were captured and transformed into web-based animation through a computation process that
joined panoramic photographs and point clouds (Salemi et al., 2005). The result supported its economic
and effective performance as being better than the computational modeling of indoor space (Salemi et al.,
2005).
2.2.2 Triangulation scanners application in Heritage Conservation
Triangulation is one of the most widely used methods to acquire spatial data from architectural features.
Scanning techniques work on triangulation including photogrammetry and structured light scanners.
2.2.2.1 Photogrammetry application in Heritage Conservation
Recent research in photogrammetry has primarily focused on monitoring changes in built heritage sites
(Liu et al., 2022; Vellanoweth et al., 2022), examining large-scale historic resources (Harbowo et al.,
2022; Simek et al., 2022), assisting visual ability under hostile conditions (Grifoni et al., 2022), and
documenting museum objects (Romano et al., 2022). Activity is growing to address the broadened uses of
photogrammetry in the field of Heritage Conservation. Photogrammetry-generated models enable the
visibility of structures and subjects at extremely close distances, and in difficult conditions for human eye
perception (Grifoni et al., 2022). In 2022, a group of scholars reviewed a wall painting using
photogrammetry on a narrow steel walkway, which restricted the view of the mural to extremely short
distances (≈40 cm) that makes general viewing difficult (Grifoni et al., 2022). Photogrammetry is also
applied in generating orthoimages of wall paintings. With a large lateral overlap ratio between abutting
shots, photogrammetry technique allows the undistorted view of the surface in hostile fruition contexts
(Grifoni et al., 2022). In the same year, the photogrammetry technique assisted researchers to examine the
relationships among glyphs and their physical contexts in ancient caves and see images that were
otherwise invisible during in-person observation (Simek et al., 2022).

45

Figure 2-8: Orthoimage of mural painting by photogrammetry (Grifoni et al., 2022).
In surveying current literature of photogrammetry scan of historic resources, scholars generated spatial
data with a variety of image sources. This source is not limited to photographs taken by professional
cameras (Attarian & Safar Ali Najar, 2022; Simek et al., 2022), crowdsourced images (Liu et al., 2022)
and videos (Vellanoweth et al., 2022). are also available for generating data with resolution that is high
enough for specific research purposes. On the upper Gulf of California, México, photogrammetry
generated from videos taken at different times was used to evaluate the erosion trend of heritage in the
coastal environment (Vellanoweth et al., 2022). The resolution has been proved largely related to the
numbers of photos taken, resolution of photographs, and lateral overlap between contiguous pictures
(Grifoni et al., 2022; Simek et al., 2022; Vellanoweth et al., 2022). Various choice of image sources for
photogrammetry also led to consideration towards cost-effective concerns in the HP domain. The historic
facade of Bothwell Castle in Britain was monitored through pictures taken by a vast number of tourists.
Scholars suggested the potential of producing small-scale digital reproduction of historic sites through
crowdsourced image photogrammetry instead of massive scale projects that increase unnecessary costs
(Liu et al., 2022). Comparison between laser scanning and photogrammetry in HC practice was carried
out; photogrammetry showed its strength in capturing data with complex geometric shapes, creating
dense and textured point clouds, and cost-effectiveness that fit better into the HP reality than professional

46

laser scanners’ heavy equipment, high cost, and lack of adequate surface coloring (Alshawabkeh et al.,
2021).

Figure 2-9 Photogrammetry from crowdsourced photography (Vellanoweth et al., 2022).

2.2.2.2 Structured Light application in Heritage Conservation

Structured Light scanners are widely applied in cultural artifacts and relics scanning. Recent literatures
reviewed its capability in historic fabric (Montusiewicz et al., 2021), artworks (Sánchez-Jiménez et al.,
2019), artifacts (McPherron et al., 2009; Rocchini et al., 2001) and architectural features (Arias et al.,
2005; Patrucco et al., 2019). Back in 2001, researchers developed a low-cost scanner with a Structured
Light system fulfilling HP interests in completed shape and requirements on accuracy (Rocchini et al.,
2001). Its valuation of different patterns, scanner calibration, and color acquisition ability were

47

thoroughly studied and examined; the result shows the limited range for a single scan led to longer
scanning time (Rocchini et al., 2001), which was also noted in later scholarly work at two Middle
Paleolithic sites in southwest France (McPherron et al., 2009). The range covered for a single scan is
determined by light emitter’s range and acquisition volume of the receiver. Even multi-station scans can
be registered to acquire large-scale objects or scenes (Shao et al., 2019), recent research with the
application of Structured Light scanners exhibited a trend on relatively small objects compared to village,
district, and city scale scanning.

Figure 2-10: Structured light scanning on the Minerva case study (Rocchini et al., 2001).

In the most recent studies, scholars used structured light scanners in developing measurement system
assisting cultural relics packaging process (Shao et al., 2019), exhibiting historical cloth (Montusiewicz et
al., 2021), and testing its precision on oil painting (Sánchez-Jiménez et al., 2019). Structured light
scanners’ limitation in small scanning range and multi-station scan was overcame through overlaid with
one TSL scan of the entire grotto. The combination saves time and energy at the same time and provides
accurate and detailed spatial data (Shao et al., 2019). The improvement and study of Structured light
scanners surround the cost-effective concerns (Rocchini et al., 2001; She et al., 2022).

48

2.2.3 Light Detecting and Ranging (LiDAR) application in Heritage Conservation
LiDAR is an acronym for “light detection and ranging.” It works by calculating distance from the time
difference between the laser beam being shot out and received. Pulse scanners and time of flight scanners
work in the same way. LiDAR scanners are the most commonly used 3D scanners for their ability to
acquire data in one scan, especially for large scale projects (Shao et al., 2019; Vavrouchová et al., 2022).
The working principle of emitting a single pulse of laser beam with strong energy enables LiDAR
scanners to acquire data from great distances. Equipped on a drone, LiDAR scanners complete the
scanning of a village from the air (Vavrouchová et al., 2022). Researchers studied the depopulation and
abandonment of rural mountain villages in post-World War I Europe (Vavrouchová et al., 2022).
Ownership boundaries of land can be clearly recognized through the LiDAR point clouds; large remnants
of buildings were detected not as ground but as buildings in LiDAR-derived DEM systems
(Vavrouchová et al., 2022). Researchers compared LiDAR-scanned villages to archival cadastral maps
and field survey, and concluded that it is the best choice in “detecting ancient ploughing patterns,
concealed under both tree canopy and turf.” (Vavrouchová et al., 2022)
In the field of historic preservation, LiDAR system has also been used for reconstructing heritage sites
and artifacts digitally (Bent et al., 2022; Shang & Wang, 2022), documenting (Yastikli, 2007), cultural
resource management (Daly et al., 2022), improving public education experience (Ballarin et al., 2018),
and evaluating technical condition of buildings (Nowak et al., 2020; Özeren & Korumaz, 2021). Ground-
based LiDAR systems, also known as Terrestrial Laser Scan (TLS) shows usefulness in building
diagnostics (Nowak et al., 2020). Researchers obtained geometrical data of a whole building including
staircases and basement with Faro Focus M70. Wall distortion and large floor deflections were diagnosed
by analyzing point cloud data and drawings. Nowak, et al., concluded that TSL scans can effectively
assist diagnostics of physical conditions, and determine the cause of damage of a building (Nowak et al.,
2020). In Özeren and Korumaz’s 2021 study, point clouds acquired from a structure by a Faro S120 Laser
Scanner were further processed and analyzed with registral and design software into HBIM (Özeren &

49

Korumaz, 2021). LiDAR and HBIM are proven to make valuable contributions to historic preservation
decision-making (Özeren & Korumaz, 2021).

Figure 2-11: TSL scanned data in Autocad measuring deformation of front wall (Nowak et al., 2020).
However, LiDAR scanners have certain limitations in unfavorable weather conditions, vibrations, and
reflective surfaces (Filgueira et al., 2017; Nowak et al., 2020). Its poor performance in rainy conditions
has been demonstrated by a variation up to 20 cm in the worst situation. Reflective surfaces, translucent
and opaque objects also cause false or missing points in the scan (Haddad, 2011). Scan data can be
improved through covering the surface with a thin layer of white powder with an approximately 45-
degree angle (Alshawabkeh et al., 2021; Haddad, 2011). Recent studies on Longmen Grottoes explored
the possibility of overlaying a TSL scan with a Structured Light scan to reach high accuracy, large range,
and high scanning speed at the same time (Shao et al., 2019).
Stating the concern on registration process, high-cost, and large equipment of TSL scanners,
(Alshawabkeh et al., 2021), scholars’ interest in LiDAR’s application in HP is leaning towards utilizing
cost-effective alternatives (Gonçalves et al., 2019; Murtiyoso et al., 2021). A first assessment of

50

smartphone LiDAR in the historic preservation domain was tested in 2021. Researchers examined the
solid-state LiDAR (SSL) embedded in Apple products in three case studies, comparing scanning results
with TSL and DSLR photogrammetry data. It proves that SSL is capable of scanning for 3D visualization,
AR, VR, while unsuitable for tasks which require higher precision such as detailed 3D printing, digital
twins, HBIM, orthophoto, texture analysis, and mesh analysis. The authors also pointed out that future
research on assessing SSL in large selection of sample objects should be carried out (Murtiyoso et al.,
2021).
2.2.4 Phase scanners application in Heritage Conservation
A large number of existing studies in the broader literature have examined phase-comparison scanners'
ability to obtain dense point cloud data (Fais et al., 2019; Fais et al., 2017) and, more specifically, detect
deficiency of surface and complex objects (Fais et al., 2017). It was reported in a study of the complex
shape of some artifacts from the “Palazzo di Città” monumental compound that long-range phase shift
terrestrial laser scanners (Leica HDS-6200 TLS) are able to generate an extremely high density of point
clouds with multiple scans (Fais et al., 2017). Determined by its working principle, Phase-Comparison
scanners (PS) exhibited strength in detecting defects in the surfaces of walls (Suchocki, 2020). It has been
used with ultrasonic tomography for detecting internal defects and heterogeneity of a comenditic
pyroclastic rock and Pietra Forte carbonate rock samples (Fais et al., 2019).

PS scanners work on emitting laser lights that are modulated in specific waveforms. Once laser light
reached the surface of the object, the intensity pattern is displaced by the impact on the surface of the
object. Measuring the differences between the outgoing and receiving laser signals provide precise
distance calculations. Recent PS scanners are proven to provide sufficient accuracy under the condition of
an angle exceeding 70 degrees with approximately 80% data loss (Mill, 2020), which is considered
accurate, fast, and provides high-resolution data (Faro, 2022c). At the same time as achieving fast and
denser data set, phase-based scanner is noisier with limited range. Limited by its working principle, PS

51

scanners are strongly impacted by tree canopy, and reduced accuracy in more dynamic range (Geo Week
News Staff, 2004).
2.2.5 360 Degree photograph application in Heritage Conservation
360-degree photography is generated from photographs taken at one location from different angles
(Panoraven, 2021). It provides a full spherical view of an observer standing at one point looking in
different directions (Panoraven, 2021). The 360-degree photographic technique was practiced with laser
scanning technology in supporting conservation design in Portugal and a Romanesque church in Spain
(Masciotta et al., 2023). Heritage conservation groups also make use of 360-degree photographs for
representation.

Figure 2-12: 360 degree panoramic photographs linked to BIM for heritage conservation (Masciotta et al.,
2023)

A series of 360-degree photographs can be joined in a 360-photograph virtual tour. In a virtual tour,
instead of viewing from a single standpoint, users can move between different photo shoot locations and
experience space. (Fig. 2-4). A past web-based research had been done at the Municipal Baths of

52

Strasbourg. Scholars took series of photographs and stitched them into 360-degree views of various
location within the baths, an interface was then created allowing audiences to pass by using a transition
from one panoramic image to the next one (Koehl et al., 2013). In-depth workflow of planning and
creating virtual tours were discussed in literatures focus on cultural heritages. The research for the historic
centre of Rethymno virtual tour creation brought a 360-degree photographic virtual tour into a game
engine and Virtual Reality to increase its immersive interactivity (Argyriou et al., 2020). 3D visualization
of cultural heritage in Caceres, Spain, combined 360-degree photographs and TSL scanning into a
hypermedia atlas through a web-page. Point clouds acquired by TSL scanners were used to make up for
360-degree photographs’ limitation in spatial data, increasing the data obtained in this hypermedia atlas
(Naranjo et al., 2018).

Figure 2-4: Michael White Adobe presented by AQYER (A Friends of the Michael White Adobe, 2022)

53

2.3 Studies using a smartphone for scanning
With a LiDAR system attached to smartphones, scanning with portable devices is an accessible approach
for an average person. Serving heritage conservation objectives with smartphone devices is of high
interest for its speed, portability, and cost-effective considerations which are not easy to meet with high-
end scanners (Spreafico et al., 2021). Past research suggests that the LiDAR system embedded in iPhone
products are solid state LiDAR (SSL), which “creates a fine grid of points, with the distance to each point
measured individually"(Murtiyoso et al., 2021). The spatial tracking ability of Apple products was
evaluated for the purpose of Augmented Reality (AR). At a 50 m distance, Apple products had a precision
at around 1.2 m, and accuracy at around 1.8 m-1.9 m range (vGIS, 2020). Further research on Apple
product scanners tested the ability of 3D rapid mapping and its quality (Spreafico et al., 2021). The
research indicated that the SiteScape app constrains the maximum size of the scanning file, which led
either to a longer scan with a lower density or increasing density limited to a smaller area (Spreafico et
al., 2021). The accuracy of tested data were compared using the ICP algorithm in Leica Cyclone 3DR
(Spreafico et al., 2021). Cloud-to-cloud distances analysis suggested that edited iPad Pro scans can reach
67% of points within 2 cm from the ground truth scan by the TSL scanners (Spreafico et al., 2021). The
evaluation led to a conclusion that point clouds obtained from Apple products with LiDAR sensors are
suitable for 1:200 map scale based on Italian standards (Spreafico et al., 2021).

54

Figure 2-12: Data comparison with TLS, DSLR, and SSL (smartphone) (Spreafico et al., 2021).
2.4 Existing research on Richard and Dion Neutra’s Reunion House
Built in 1951, the Reunion House was programmed and designed as a place where grandparents and
grandchildren can sleep separately but share common living spaces. The house underwent a series of
alterations based on Richard Neutra’s son Dion Neutra’s needs when lived on the property. The Reunion
House was listed as a City of Los Angeles’ Historic Cultural Monument in 2021.

Character-defining features (CDFs) contribute to the significance of the house; they are also the key
elements to be documented and conserved within the whole context of the building. Exterior and interior
CDFs are identified in the designation application. In addition to documents provided available from the
City of Los Angeles, a group of students from University of Southern California produced an assessment
and recommendations report and presentation for the property in 2021.

55

2.4.1 Reunion House developmental history (Lamprecht, 2020)
Arthur L. Johnson, Jr. commissioned Richard Neutra to build a single-family house after he purchased the
lot at 2442 Neutra Place on Sept. 27, 1949. A permit was issued from the City of Los Angeles
Department of Building and Safety for a “New building permit for home 34’ x 91’” after a year of design.
In the following year 1951, Johnson Jr. received his Certificate of Occupancy and was able to move into
the property. The Johnsons, sold the property to Alphonse D. Makowski and Ann L. Makowski. in the
following year. On May 2, 1962. The house owners sold the property to William Hobson and Evelyn T.
Hobson. In 1963 Dion Neutra bought the property from William Hobson and Evelyn T. Hobson and
began a series of alterations based on his understanding and needs. A reflecting pool adjacent to the west
elevation of the building was added in 1964; interior features such as shelving and mirrors next to the
fireplace, as well as the kitchen curtain were added at the same time. In 1966, the kitchen was remodeled
and brown-stained concrete in the living room was carpeted. Additional alterations including the exterior
walls, interior lighting systems, and restoration of the ceiling to the original wood finish had been
completed by Dion Neutra between 1966 to 1968. In the bedrooms, Dion Neutra added shelving, a closet,
a desk unit, and lighting according to his needs. Behind the master bedroom, Dion Neutra added a closet
addition for his wife. In 1968, Dion Neutra decided to add a second floor to the garage based his father,
Richard Neutra’s original drawing for Arthur Johnson. In order to accomplish it, Dion Neutra reinforced
the structure to support the unit above the garage and added a driveway for the renter’s vehicle. At the
rear of the house, a retaining wall was also added. These alterations are considered significant because
they were designed by Richard Neutra’s son, Dion Neutra during the period when the house took the form
that exists today.
2.4.2 Reunion House Character Defining Features
As a case study example for the smartphone’s ability to meet heritage conservation objective, Reunion
House’s Character Defining Features (CDFs) are key elements to be scanned within the whole building
entity (Table 2-1).

56

57

Table 2-1 : Exterior Character-Defining Features (Lamprecht, 2020)
Exterior Character Defining Features –
Mid-Century Modernism
Exterior Character Defining Features – Neutra
a long, horizontal profile reinforced with a flat
roof
use of stucco walls contrasted with casement and fixed
windows and sliding window walls, to effect an
aesthetic of alternating solids and voids
a deep integration with site, setting, and
landscape through extended overhangs copious
amounts of glass materials that continue from
inside to outside, bridging interior and exterior
use of paint – white, dark brown, and here, silver
(common to Neutra’s window frames, posts, and sills)
and grey. These colors were used in order to project
(white) or suppress them or make them recede (brown.)
Based on Gestalt aesthetics, this is an additional strategy
specific to Neutra to introduce another kind of “solid-
void” relationship. Silver (actually aluminum) paint was
used both to protect rust-prone steel and to
“dematerialize” window frames or his 4”x4” wood posts
for a more uninterrupted view to nature, based on
Neutra’s knowledge of evolutionary biology and the
African savannah.
post-and-beam construction, or the regular
disposition of posts
diagonal views through mitered glass corners
or through simple, minimal vertical member at
corner
windows usually sliding, casement, jalousie,
or fixed lights, with simple frames that appear
commercial in origin
projecting beams extending beyond the building
envelope, either floating free, or terminating in a post as
a “spider leg”
doors are usually single-panel wood or
painted, with no ornamentation or elaborate
detail
deep overhangs, often with strip lighting flush with
overhang and at its edge. rounded post caps, created by
adding a separate piece of lumber, flat on one side and
subtly rounded on the other, which fit over a squared 4’
– 4”, thus softening the visual effect of an otherwise
rectilinear composition
use of simple, modern materials: concrete,
stucco, float glass, steel, and aluminum,
contrasted with natural materials such as brick
and stone, either random or ashlar cut
a rhythmic distribution of details, wall
treatments, textures, and windows. lack of
applied ornament
reflecting pools adjacent to the house to reflect nature
post

58

2.4.3 Documentation techniques used previously on Reunion House
The Reunion House is currently under Neutra Institute ownership and stewardship. Past scholars studied
the history of the house from its past to today. City survey documents, archival photographs, interviews
with Dion and Richard Neutra, site maps, periodicals, books, and various materials are available for study
and reference. The documents are mostly hand drawn design and construction documents on paper or film
based.

In 2021, Reunion House was listed as a City of Los Angeles Historic-Cultural Monument. The official
agenda package for designation includes the Final Determination Staff Recommendation Report,
Categorical Exemption, Under Consideration Staff Recommendation Report, and Historic-Cultural
Monument Application. Brief introduction, general information, and California Environmental Quality
Act (CEQA) findings are discussed in Final Determination Staff Recommendation Report. The report
determined that the Reunion House meets criteria 1) be identified with a significant event, 2) be
associated with important people, 3) embodies the distinctive characteristics of a style. In the report, an
architectural description is included. The piece of writing detail documented the building’s relationship to
the tract, direction, size, and major components of the structure (see Appendix B) (Lamprecht, 2020).
Then each elevation was elaborately described. In Lamprecht’s report, the physical appearance and the
notion of Neutra’s architectural design was pictured through written expression. While A verbal
description of a building is simply not enough to properly capture the essence of the structure. A 3D scan
of the building is necessary to accurately represent its size, scale, and dimensions, as well as its three-
dimensional qualities such as its structure, design, and composition. A 3D scan can be used to document a
building or structure in ways that words alone cannot achieve. It captures the three-dimensional quality of
architecture and space, allowing viewers to see the building from different angles and perspectives. A 3D
scan can also provide the exact dimensions and scale of a building, providing a clear idea of its size and
layout. In addition, scans can be used to document the building in case of its destruction or loss. It can

59

also be used to track changes in a building over time, allowing for an examination of the evolution of a
structure. Finally, 3D scans can be a powerful tool for conveying emotion and meaning, allowing viewers
to understand the significance of a building even if they have never seen it in person.

Shortly after the Reunion House was listed as the City of Los Angeles Historic-Cultural Monument, a
group of University of Southern California Master of Heritage Conservation students studied the house
and produced an assessment and recommendation report for Reunion House (National Park Service,
2022a). The report was organized with each feature with a caption number, photograph(s), caption,
estimated date, feature type, significance, description, condition class, condition description, and
recommendation for treatment. Photographs, description, and condition description mainly surround the
geometrical shape, material, and physical condition of features. A photograph next to the written
description of an architectural feature can provide direct information about the feature itself, such as its
size, shape, color, and materials. However, it is limited in its ability to relate individual features to the
larger entity of the house. For instance, a photograph of a window may not provide enough information to
determine how the window relates to the overall design of the house, such as its position within the
overall floor plan, or how it contributes to the overall aesthetics.

A technical plan needs to be established to bridge the gap between text and perception, segment and
totality. To do this, the following features and elements should be taken into account for each task related
to Heritage Conservation:

60

Table 2-2: Scanning focus of each heritage conservation tasks

Site Survey Documentation Monitoring Education
Interior Wall
projecting beams extending beyond the building
envelope, either floating free, or terminating in a post as
a “spider leg”
Exposed post
and beam
structure
Interior
decoration,
staging
Exterior
Wall
Deep overhangs, often with strip lighting flush with
overhang and at its edge. rounded post caps, created by
adding a separate piece of lumber, flat on one side and
subtly rounded on the other, which fit over a squared 4’ –
4”, thus softening the visual effect of an otherwise
rectilinear composition
Wall
Exterior,
landscape,
public road
Door,
window, and
opening
location
A rhythmic distribution of details, wall treatments,
textures, and windows
Ceiling
Lot terrain Floor

2.5 Summary
This chapter described 2.1 3D scanning’s application in general fields of industry, 2.2 3D scanning’s
application in the field of Heritage Conservation, categorized with different working principles, 2.3 past
studies using a smartphone for scanning, 2.4 past research and documentation done for the Reunion
House, and 2.5 summary.

Strength and limitations of various professional scanners are thoroughly studied. They are more than
capable of documenting and providing useful information for future design and conservation work.
Different types of scanners exhibited their distinctive vantage points and shows their limits. Studies and
technologies work on offset these limitations and result in higher quality of scanned data.

Triangulation, pulse, and phase-comparison scanners are the most commonly used technologies in the
field of heritage conservation. These scanners produce digital data which serves as the basis of any
preservation project. With the rise of more affordable and portable methods such as smartphones
equipped with LiDAR sensors, more researchers are investigating the accuracy of these scanners. While

61

much work has been done in this area, further studies need to be conducted to determine which method is
best suited for which type of heritage conservation project. This would help both professionals and
amateurs acquire the necessary data for their projects.

The Reunion House at Silver Lake was selected as a case study. Listed as City of LA Historic-Cultural
Monument in 2021, Barbara Lamprecht and Neutra Institute developed a comprehensive document
including survey, historic context statement and architecture description. In the same year a group of
students, from USC heritage conservation program conducted an assessment and recommendation for the
property. The package of documents offers a wealth of information, both tangible and intangible, that can
be used as a basis for further research on the use of smartphone-scanned data. In addition, there is
potential to incorporate newer, more cost-effective technologies such as scanning to enhance the
document package, which could be applied to areas such as site surveys, monitoring, and education in the
field of heritage conservation (Figure 2-14).

Figure 2-14: Smartphone scanning technologies match with heritage conservation goals

62

Chapter 3 Methodology
Chapter 3 will discuss the scanning device and software using for scanning; site survey for heritage
conservation using smartphone, including software selection, data acquisition, data processing,
comparison to traditional methods, and case study done by professional scanners; documentation for
heritage conservation; monitoring for heritage conservation; and education. Chapter 4 will discuss all the
steps in more detail using the test scans as examples.

Scanning has been used in various aspects in the field of heritage conservation. The high expense and
sophisticated operation of a professional scan can sometimes be infeasible with respect to cost and
inaccessibility of sites. An alternative scanning method is proposed using a portable smartphone equipped
with a high-resolution camera and LiDAR sensors. Smartphones can capture 2D photographs, cylindrical
panoramic photos, 3D Matterport pictures, photogrammetry, and some of them can even generate LiDAR
point clouds. Two case studies are proposed: a partial study of the Hoose Library of Philosophy at USC
and specific areas at Richard Neutra’s Reunion House in Los Angeles, CA. HC tasks, namely site survey,
documentation, monitoring, and education, will be practiced with the proposed method. The methodology
will emphasize distinctive purposes for each task. An iPhone 13 Pro will be used to examine the
performance with selected iOS apps. Additional free software will also be applied in processing data. The
results of the test scans at Hoose Library and the case studies at Richard Neutra’s Reunion House will be
compared respectively with traditional methods and high-end scanners.

63

Figure 3-1 Proposed methodology

64

3.1 iPhone 13 Pro as scanning device
The LiDAR system was first installed on Apple products in 2020 with the release of iPhone 12 Pro and
iPad Pro (4th generation). In the very next year, Apple released their iPhone 13 Pro and iPad Pro (5th
generation), which are also equipped with LiDAR systems. Even though released in different series, the
LiDAR technology and equipment used in all Apple products are the same (Stein, 2022). The differences
among these smart devices lie in the size of screen, storage memory, and processing chips.

The proposed method will be executed with an Apple iPhone 13 Pro, released in September 2021, priced.
It has three Pro 12MP rear cameras, telephoto, wide, and ultra-wide cameras - and a LiDAR sensor, with
a total weight of 203 grams (7.16 ounces). With its height of 146.7 mm (5.78 inches), a width of 71.5
mm (2.82 inches), and a depth of 7.65 mm (0.30 inch), iPhone 13 pro is portable and lightweight. With
cost-effectiveness being the concern, the price of the iPhone 13 Pro is US dollar 999.99 plus tax (as of
October 2022).

The LiDAR sensor embedded in Apple products is claimed to be using ToF technology, which is
considered solid-state LiDAR (SSL) as they do not have to move parts for a scan (Murtiyoso et al., 2021).
The LiDAR scanner is responsible for measuring the distance from the device to objects in its vicinity,
while the projector sends out infrared light. The image sensor captures the light that is reflected off of the
objects around the device, and the processor then uses this information to create a point cloud file of the
surrounding environment.

65

Figure 3-2 Using smartphone scan a cultural heritage (Crabbe, 2013)

3.1.1 Scanning app: SiteScape and Matterport
SiteScape is a software program that uses LiDAR scans for architecture, engineering, and Cconstruction
industries. It has a user-friendly interface; after signing up for free with name, email, and password, a
camera page would show directly (Figure 3-3). There are three buttons at the bottom of the interface;
from left to right are customizable acquisition settings, tap to start, and album. The app allows
customization of “Point Density” (“low”, “medium”, or “high”), and “Point Size” (“low”, “medium”, or
“high”) (Figure 3-4). The “point density” option can change the number of points captured; the option of
“medium or high” can capture two or four times the number acquired in “low” quality mode, which
impacts processing time and produced data (need a citation here). The “point size” option only affects the
dimension of dots visible on the interface, which does not impact the obtaining of data.

66

Figure 3-3: SiteScape iOS app Figure 3-4: SiteScape parameter

The iOS application SiteScape version 1.6.5 by SiteScape Inc. is then chosen as it meets specific needs
related to cultural heritage documentation: 1) the app can is free to download and use (upgrade option is
available for more functions) 2) acquisition settings can be easily adjusted upon requirements 3) 3D
models can be generated as a point cloud. These attributes allow an average person without professional
experience to operate the scan with no cost and generate the final product with various raw data sets.

The Matterport company released Matterport Capture app in 2020 to bring 3D data acquisition to the
iPhone. In both scans at the HLP and Reunion House, version 5.2 of Matterport Capture app will be used.
The application comes with free user licenses and the accounts that are accessible for scanning practice.
Its easy-to-use interface allows people in the field of historic preservation to use it as a tool assisting their

67

projects. Matterport Capture has multiple scan options including scan types of “3D Scan” and “360
Capture” with scan modes of “LiDAR Scan”, “LiDAR Complete Scan”, “Simple Scan”, and “Complete
Scan” (Figure 3-5). According to Matterport. Inc, “the 360 Capture will generate a 360° view of a space.
A 3D Scan will generate the data needed to create a 3D model or virtual tour of a space.

Figure 3-5: Matterport Capture iOS app (Matterport, 2022)
3.2 HP survey with smartphone
A survey allows for a systematical identification and record of heritage resources in the community. Site
surveys are labor intense work that it requires researchers go out to the field and record buildings and
terrains on site. The information documented by researchers on site will be significant in creating a
foundation for the joint of physical condition of the site and archival knowledge. A site survey will be
created through smartphone scanned data and compared with traditional map and floorplan techniques, as
well as a site survey conducted with high-end scanners.

68

3.2.1 Software selection – SiteScape
SiteScape is selected as the smartphone phone app used for generating heritage conservation site surveys,
because the application allows for pause during a scan. Different density of scans gives more options to
researchers. A high density scan might be used if you needed to capture a lot of detail in a short amount
of time. For example, when creating a map of a small area, a high-density scan could capture the space
with more points describing it. A low-density scan might be used if you needed to capture a large area but
did not need a lot of detail. For example, if you were creating a map of an entire building, a low-density
scan would be a better choice as it would capture the area in thin layer of points and allows for larger
scanned area.
3.2.2 Data acquisition process
In the data acquisition process, SiteScape encourages a distance between 3-12 feet (1-4 meters) from the
scanned objects. To capture quality data, the scan is needed to be executed in a smooth path, avoiding fast
and large movement, especially in horizontal direction. SiteScape noted that to scan the outdoor
environment, direct sunlight should be avoided, cloudy days, post dawn or before dusk are more suitable
conditions. Test scans will be implemented at the Hoose Library of Philosophy before the case study scan
at Reunion House being conducted. For the convenience of the tenant of Reunion House and the
researcher, scans will be arranged in afternoons. Scans of interior space will be performed prior to the
garden, with a north to south order depending on the location and lighting condition of the selected
rooms/ features. Each scanning object will be scanned at least three times with parameters of point
density with “Low”, “Medium”, and “High”. Repeat scans are encouraged when initial trials have visible
holes, double layered points, or missing areas.

69

During each scan, a timer will be set in recording the human effort in completing a scan by smartphone.
Monitoring the device’s condition will also be noted. SiteScape mentioned an overheated central
processing unit can drastically reduce scan performance, and low battery would also impact scan quality.
When the temperature of a smartphone screen or back panel is higher than an average human hand
temperature, the scan procedure shall be paused until the temperature drops back to normal. Phone battery
shall remain over 50% over the whole course of scan, data processing, and synchronizing. A portable
power bank or phone charger should be available at the scan site.

Misleading registration of geometrical data will occur when there is only one surface or texture (wall,
carpet, or grass) within the scanning frame for several seconds. This is because SiteScape automatically
registers the LiDAR sensor that the situation will cause miss-aligned points. To prevent the misleading
circumstance, scans are encouraged to keep multiple features on screen. A timer will record how long
each scan takes.

After scans complete, each point cloud will be synchronized into SiteScape cloud, initially processed, and
downloaded in three different formats (.ply, .rcs, .E57) respectively (free user license only allows one
model at a time). A hands-on and free cost software, CloudCompare, will be downloaded and used for
data processing.

The process remains the same for test scans and the case study.
3.2.3 CloudCompare: merging multiple scans
The free desktop software CloudCompare will be used for data processing after the point cloud data is
downloaded locally (Figure 3-6). Raw data contains noises and unwanted parts, thus before any analysis
started, scan data need to be cleaned and copped. Automatic noise cleaning can be achieved using SOR

70

and Scalar fields function in CloudCompare, and cropping can be done with the scissor icon on the
toolbar.

Figure 3-6: CloudCompare interface

In order to achieving a survey, the methodology needs to use multiple point clouds in generating a large
floorplan or map. Smartphone scanned point cloud has a maximum file size, which means the app is not
capable of scanning the entire building or site with one scan. The method of scanning a space with
multiple scans, and merge them together in CloudCompare will be used. Each scan will slightly overlap
the previous scan with the assist of reference points.
3.2.4 Smartphone acquisition qualification

71

Smartphone acquisition result for survey will be evaluated based on its feasibility, quality of result, and
limitations and strengths of the methodology. The analysis will involve considering factors such as the
cost and time required for data collection, as well as the availability of suitable equipment and personnel.
The quality of the data collected will also be accessed, including factors such as the clarity of the images,
the accuracy of the measurements, and the consistency of the data. Meanwhile, recommendation will be
made based on research result to assist heritage conservation site survey purposes.
3.3 HC documentation with smartphone
Historic preservation documentation aligns with the purpose it serves. Most generally, documentation
requires capability of recording geometric shapes, color, texture, and interrelationships between
architectural features. The iOS application will be selected and installed, parameter settings and interface
will be studied. Acquisition procedure will be performed with the device with different parameters. Two
comparisons will be made: the expense, operation, and results will be compared with 1) traditional
methods of historic preservation documentation, 2) high-end scanners application in the field.
3.3.1 Software selection – SiteScape
When documenting heritages, architecture historians, cultural resource managements, and heritage
conservationists are concerned with the amount of detail being recorded. The geometry of character
defining features and architectural features shall be accuracy described in documentation. The smartphone
software selected for documentation purpose should be capable of registering concreteness of a detail of
architectural features, including texture, shape, and color.

Beyond the competences of the application, cost and operability by non-technicians as factors should also
be considered when selecting smartphone applications. iOS scanning app SiteScape is then selected. As
introduced in 3.1.2, SiteScape is free for use, easy operation, and produce point clouds with different
parameters, which meet the needs for heritage conservation documentation.

72

3.3.2 Data acquisition process
The same procedure as 3.2.2 Data Acquisition Process will be practiced for documentation purposes,
focusing on architectural features and character defining features mentioned in Chapter 2, Table 2-1. The
process remains the same for test scans and the case study.

Table 3-1: Scanning focus for case study at Reunion House
Features to Scan Past Documentation
Interior, exterior wall, ceiling, floor original drawing
Projecting beams extending beyond the building envelope, either floating free, or
terminating in a post as a “spider leg”
photographs
Windows, doors, and their location (a door connecting kitchen to study was
changed to a bookshelf. The time of alteration was determined the same time as
the kitchen alteration, which post the period of significance)
original drawing,
photographs
Current Kitchen (Dion and Richard did multiple alterations to the kitchen,
including the cantilevered countertops in the kitchen/breakfast nook, large mirror
on the north side of the kitchen wall, and a second mirror in the southwest corner
of the kitchen. )
original drawing,
photographs
Interior decoration, staging, exterior, plant canopy photographs
Built in furniture
photographs,
drawings
3.3.3 Using CloudCompare processing data: increasing precision

73

The same automatic noise cleaning and cropping procedure as 3.2.3 will be conducted on the data
acquired from 3.3.2. Then, the goal of increasing precision will be introduced.

Precision describes how close measurements of the same item are to each other. In an accurate but not
precision situation, points will surround the geometry of object, but scattered around. To increase
precision, two scans will be registered together in testing if overlay will increase the number of points
describing the same space. The number of total points describing the defined space will then be used for
evaluating the methodology’s competence in increasing precision.

Two overlay and processing orders of the two scan data will be considered
1. Noise clean of each scan will be accomplished before the two scans being overlaid
2. Noise clean will be accomplished before the two scans being overlaid.

3.3.4 Smartphone acquisition method qualification
The point cloud data acquired with smartphone devices are evaluated on its ability to record geometric
shapes, surface textures, color, complex structures, interrelationships between different architectural
features according to American Historic Building Survey suggestions. Beyond its performance, level of
operability will also be compared with traditional methods of measured drawing, large format
photographs, and written descriptions. Whether an average person in the field of HP can or cannot operate
this method is going to be discussed. Time needed for a quality scan, potential difficulties, and basic data
processing procedures will also be discussed. Aiming at mitigating the cost-effective concern for HP
projects, the cost of a full scan will be included in the comparison as well.

74

3.4 Monitoring
How cultural heritages change overtime can be watched through monitoring its physical condition. The
condition can be recorded with 3D scanning technology. Heritage conservationists can observe the change
through comparing current scan with scans taken years ago. If a building structurally shifted, the two
scans will have distance instead of accurately overlaying each other. Thus, each scan needs to closely
measure the true or accepted value in order for researchers distinguish the nuance in structural changes.
The smartphone and desktop software selected for building and site monitoring will be explored.
Acquisition and data processing will be practiced in test scans before scanning the case study at Reunion
House. Comparisons will be made on the expense, operation, and scanning data result with 1) traditional
methods of historic preservation documentation, 2) high-end scanners application in the field.

3.4.1 Software selection – SiteScape
To monitor a building, the change of the building’s condition and geometry over time need to be
documented and contradistinguished. This could involve monitoring the construction of a new building,
the demolition of an existing one, the erosion or accumulation of land, or any changes to the existing
terrain of the building or site. The monitoring would include measuring the changes in height, width,
length, angle, and any other geometric characteristics of the building or terrain. This would be done
through the use of surveying equipment, such as total stations or drones. The data collected would then be
used to create detailed documentation of the changes and be used to inform future decisions about the
building or site. The collation between documentation taken from different times can spot changes in
material and structure.

This includes cracks, deterioration, and shifts in structure. Cracks and deteriorations can be sighted with
human eyes, but the nuance shifts in a structure are hard to observe. The iOS app chosen for monitoring

75

purposes should be able to export point clouds for further data processing. The free iOS app SiteScape
introduced in 3.1.2 is then chosen because it meets the needs to monitor a building.

3.4.2 Acquisition for HP monitoring
The same procedure as 3.2.2 Data Acquisition Process will be practiced for monitoring purposes. The
data obtained from this process will be used in CloudCompare for analysis. Before scanning, make sure
the environment is prepared for a successful 3D scan. This includes clearing the area of any potential
obstructions, ensuring adequate lighting, and setting up the smartphone app. The focus of monitoring
acquisition should focus on the structural wall, ceiling, columns, and beams (if visible).

The process remains the same for test scans and the case study.
3.4.3 Using CloudCompare processing data: increasing accuracy
Two scans of the same space of test scans (Scan 1 and Scan 2) will be used in data processing. Both scans
need to be cleaned and cropped follow instructions in 3.3.3. The accuracy of the point cloud data can be
improved with the techniques such as outlier removal and noise reduction.

The methodology of overlapping two scanned point cloud data is then developed to further increase
accuracy. Accuracy describes how close a measurement is to the true or accepted value. Thus, the
distance between the two clouds exhibits their ability to accurately describe the target. This process
involves combining two separate point clouds that have been captured from different trails to create a
more accurate and complete 3D representation of a space. The process begins by aligning the two point
clouds to each other. After the two point clouds are aligned, the next step is to calculate the differences
between the two clouds. This difference is then used to determine the accuracy of the combined point
cloud with cloud-to-cloud distance function in CloudCompare.

76

3.4.4 Smartphone acquisition qualification
Smartphone acquisition for heritage conservation monitoring purpose will be evaluated based on the
repeatability of smartphone scans. Through cloud to cloud distance computation, the differences will be
visualized and compared to professional scanner’s accuracy. Recommendations will be made based on
research results.
3.5 Education
Heritage conservation is the process of preserving and protecting the natural and cultural heritage of a
place. It involves the identification, protection, conservation, and management of places, objects, and
other cultural resources that are of special importance to a particular community. Heritage conservation in
education refers to the teaching of methods and practices of heritage conservation in school settings. This
includes lessons on history and culture, the importance of preserving and protecting the heritage of a
place, and the role of students in preserving and protecting their own cultural heritage.

Visualizing heritage conservation in education can involve a variety of different approaches, such as
creating a visual timeline that maps out the history of conservation efforts, creating infographics that
illustrate the environmental, economic, and social benefits of conservation efforts, creating interactive
maps of protected areas and monuments, and creating spatial experiences to bring heritage conservation
to life, where 3D scanned data can be helpful. The ability to visualize and provide information becomes
the main concern for this purpose. A smartphone app will be selected in obtaining spatial data and create
the experience in a cost-effective method. Its competence in delivering spatial experience and shareable
content will be compared with traditional teaching approaches and projects done with professional
scanners and crews.

77

3.5.1 Software selection – Matterport
The Matterport iOS app is selected for the purpose of heritage conservation education. As mentioned in
3.1.2, A virtual tour in the visualization of space is a computer-generated simulation of physical space.
The app is capable of obtaining spatial images and experiences with 360-degree photographs and the
virtual tour function. Matterport software become a competitive candidate for also because of its powerful
and accessible data processing engine. The product of the scan can be exported to a webpage and links,
which can be viewed by a mass audience on the internet world-widely.

3.5.2 Data acquisition process
For a one-floor interior space, the tripod can be set to the height of human eyes. After setting up the
smartphone device on the tripod, one can start the scan by clicking the capture button. Matterport Capture
will provide instructions to point the phone camera at a series of white dots (Figure 3-7). During one
capture, the process can be tracked by looking at the pink rings around each of the dots. After 360-degree
rotation is completed, the tripod and device shall be moved to the abutting scanning spot, which
according to Matterport Help Center, is around 5-8 feet (1.5-2 meters) from the previous scan. The same
scanning process should be practiced at each scan spot. At least three scans should be done to complete
one survey. 360 Capture relies on AI technology in calculating depth, which is not interfering with strong
light, while LiDAR scan will be affected by such lighting conditions. Thus, during a scan with LiDAR
mode, strong light should be avoided.

78

Figure 3-7: Matterport Capture interface on iPhone

3.5.3 Using Matterport webpage processing scanned data
The obtained data will be synchronized to Matterport Cloud, where data can be viewed and processed
online through a desktop (Figure 3-8). Matterport webpage allows user to upload and edit their capture of
space. With a free user license, the cloud only allows one scan to be processed at a time. A link can be
created for public viewing of the 360 degree photograph virtual tour. Each scan should be saved as a
shareable link, then the next scan can be synchronized and processed. After all scans been processed, a
comparison among each trial with the link should be conducted, a best performance will then be selected.
The virtual tour can also be implemented into websites and shared.

79

Figure 3-8: Matterport webpage editing interface

3.5.4 Smartphone acquisition qualification
Smartphone captured virtual tour will be compared to virtual tours available online. Deliverables and
downloadable files will be exported to Matterport project webpage and manipulated; test including
dollhouse view, measurements, and virtual tours.
3.6 Summary
Chapter 3 discussed the scanning device and software using for scanning; site survey for heritage
conservation using smartphone, including software selection, data acquisition, data processing,
comparison to traditional methods, and case study done by professional scanners; documentation for
heritage conservation; monitoring for heritage conservation; and education.

80

Following this methodology, one can use a smartphone and select software to scan a heritage site in
reaching HC goals of surveying, documenting, monitoring, and educating. Matterport and SiteScape
smartphone apps were chosen for their outstanding capability in meeting these research goals. SiteScape
was designed to use in data acquisition process of intentions including surveying, documenting, and
monitoring. Computer desktop software CloudCompare would be used for registering multiple scans,
increasing accuracy and precision. Smartphone scanned data would then be compared with traditional
methods and a professional scanner scanned data. Matterport was selected for the goal of education, the
data acquisition, processing, and visualization would be juxtaposed to traditional method and a high-end
scan.

Chapter 4 will describe in further detail the methodology described using the test scans as an example.
Chapter 5 will be conducted at the Reunion House as a case study.

81

Chapter 4 Test Scans Using the Smartphone
Chapter 4 describes test scans for different heritage conservation intentions at the Hoose Library of
Philosophy at the University of Southern California. In Chapter 4, a step-by-step detailed instructions for
using the app and software operation are provided. In addition, preliminary results of test scans are given
at the end of the chapter. The comparison to traditional methods and professionally scanned data will be
carried out in Chapter 5 as a case study.

Chapter 4 is a start-up guide for scanning acquisition and data process. Scanning and processing for the
four heritage conservation purposes will be taught step by step in 4.1 Smartphone scan for heritage
conservation documentation, 4.2 smartphone scan for heritage conservation survey, 4.3 smartphone scan
for heritage conservation monitoring, 4.4 smartphone scan for heritage conservation education (Figure 4-
1).

82

Figure 4-1: Chapter 4 content diagram

83

In each of these sectors, test scans will be conducted with different approaches and software in achieving
distinct goals (Table 4-1).

Table 4-1: Test scan tasks of Hoose Library for heritage conservation purposes.

Prior to scanning, the smartphone should be fully charged and the software SiteScape and Matterport
Capture installed. The desktop processing software CloudCompare should be also downloaded. A tripod
is recommended, if not, a long stick or post can also assist the work. Other than the uploading and
downloading process, internet connectivity is not required for data acquiring procedures.

SiteScape and Matterport Capture(phone app) can be downloaded from Apple Store.
SiteScape project page can be accessed from https://app.sitescape.ai/projects
Matterport Cloud can be accessed from https://my.matterport.com
CloudCompare (desktop computer) can be downloaded at https://www.cloudcompare.org

84

4.1 Smartphone scanning overview and directions for documentation
For the purpose of documentation of cultural heritages, scanning with smartphone will be conducted with
steps introduced in the following “getting start guide.” The goal of this section is to validate the precision
of different scanning modes.

Each of the multiple scans for an HC documentation of an environment followed these steps. To scan an
enclosed space, the researcher should start facing one corner of the space, with a distance of 3-12 feet to
the scanned object. After setting up the proper parameter, the researcher should hold the smartphone
device vertically with both hands in a position where arms next to the body tightly, and ensure the screen
is clearly in sight. After pressing the start button, the researcher should use the scanned object as a center,
slowly move left and right on foot. After one feature is captured from left, right, up, and down, the
researcher should gently move to the abutting features and scan. In the case of the test scan, the researcher
should face the target surface, start scanning from right to left (southwest to northeast). The person can
move arms up to down, right to left in capturing more detail of the bookshelves (the example used in
4.1.1). When the first shelf is scanned, research should step left with device sensor facing the shelf and
always ensure there is no gap or hole in scanned data.

As the scan starts, the camera view will be disabled; when moving around the scanned object, captured
points will gradually show up on black screen so holes from data can be visible. Overlap layers of data
can be misleading and inaccurate; to prevent the situation, researcher should avoid scanning the same area
twice. During the scanning process, researcher need to scan continuous surfaces, with rich features in
sensor sight. For the purpose of generating maps and locations of walls and features, plane surfaces
should be avoided. As much as possible features and details should be included. The point count bar at
the bottom of the screen shows the total points captured; scanning will stop when the bar is full. A single
scan has limited allowed point, more scans might be needed for the entire space; the researcher should

85

take as much scans as possible to fully document the space. Each scan should overlap the previous scan
for data register. After a scan is accomplished, SiteScape allows scans to be synchronized into the clouds,
where users can view it on a desktop with a web page. With a free user license, one can synchronize one
point cloud at a time.

4.1.1 Data acquisition
Data acquisition of smartphone scanned heritage conservation documentation including major steps of
software preparation, data acquisition and exports. The smartphone application in use is SiteScape.

I. Preparation
a) Havethe smartphone application SiteScape downloaded to the device, and the software
CloudCompare installed to a desktop computer.

b) Open SiteScape, sign up a free account (Figures 4-2 and 4-3).

Figures 4-2 and 4-3: SiteScape register interface

86

c) Remove unwanted furniture or objects from the space to be scanned. The examples shown are from
the Hoose Library of Philosophy (Figure 4-4).

Figure 4-4: Hoose Library of Philosophy

d) Click on setting button, change point density to Med and Point size Low, then click Close (Figures 4-
5, 4-6, and 4-7).

87

Figure 4-5, 4-6, and 4-7: SiteScape setting button and scan setting

II. Data Acquisition
Smartphone scanned data can be acquired with or without the assist with a tripod. The two methods are
introduced here.

Method A: Without Tripod (If equipped with a tripod, see method B)
a) Hold smartphone device vertically with both hands, arms next to the body tightly. Stand 5 feet away
from the scanning object to ensure that the smartphone camera can capture rich and detail features

b) Tap the circle button to start scanning. Once started, the camera view will be turned off (Figures 4-9
and 4-10).

88

Figures 4-9 and 4-10: SiteScape operation interface and camera view adjust

c) Set the scanned object in the center; slowly move left and right on foot. Capture the object from left,
right, up, and down with colored points on screen fully describe it.

d) Gently move to abutting features and scan repeating step 6, avoiding features from reappearing on
screen.

e) During scan, pay attention to process circle at the bottom of the screen, when process bar is full, a
scan is automatically completed (Figure 4-11). A scan can be paused and resumed by clicking the
circle. If the scanning is not going well, one can restart (Figure 4-12).

89

Figure 4-11 and 4-12: SiteScape pause and resume scanning button

f) The scan will automatically complete when the process circle full. To complete a scan when the
circle is not full, click pause, then the complete button. Acquired data will be shown in a virtual
space from screen; the user can choose to Close, Export, or Sync to Cloud (Figure 4-13, 4-14). Click
on Continue & Replace if using free account (Figure 4-15).

90

Figure 4-13, 4-14, and 4-15: SiteScape post scan exporting and synchronizing

g) Repeat process 2. a) – f) on abutting structures until cover the entire interior and exterior.

Method B: With Tripod
a) Select a tripod standing point; keep at least 3 feet away from the scanning target. Stable the
smartphone on the tripod.

b) With the smartphone vertical, tap circle button to start scanning. Once started, the camara view will
be turned off (Figure 4-16 and 4-17).

91

Figure 4-16 and 4-17: SiteScape operation interface and camera view adjust

c) Holding the smartphone, steadily rotate 360 degree to capture a horizontal loop of the room. Watch
for points captured at the beginning of the scan, reduce overlapping those points when stop. During
the acquisition process, move with the camera to avoid being captured in the scan. After a horizontal
360 degree scan finished, click circle button, then complete.

d) Change the angle of the smartphone; this can be achieved through using high-end tripod with a fluid
head, or a regular tripod with carful operation. A 360 rotation capture has to be conducted twice, one
with the smartphone camera taking an approximate 30% angle facing up, and one with the camera
with an approximate 30% angle facing down (Figure 4-18, 4-19, and 4-20).

92

Figure 4-18: The smartphone device on tripod with an angle facing down

Figure 4-19 and 4-20: Tripod angled up and down

e) Repeat process Method B a) – d) on abutting structures until cover the entire interior and exterior.
Tripod standing points should have overlapped area for registration process. In narrow space, the
standing points should be closer to each other.

III. Save and Export
After finishing the scan, researchers should rename, sav,e and export files to usable format so further
data process can be done.

93

a) Select the scan model from the library, click on the Option button, and Rename the scan data
(Figures 4-21, 4-22, and 4-23).

Figures 4-21, 4-22, and 4-23: SiteScape library and renaming file

b) Move scanned data around, zoom in and out with finger, check for holes, missing parts, and double
layers (Figures 4-24 and 4-25). A double layer is two layers of points describing the same surface
(Figure 4-26). The two (or multiple) layers are created when an object is scanned or appeared twice
in scanning range. Even though the layers of points are describing the same surface, the geometry
might be accurate and repeated, the location of layers can various. The same information in different
location causes data to be inaccurate and unrecognizable.

94

Figures 4-24 and 4-25: Zoom in and out scan of Hoose Library of Philosophy on SiteScape

Figure 4-26: Double layer of points in scan 2

c) If any miscapture is detected in b), improve the scan path and repeat step 2. A) to 3. B) to rescan.

95

h) Click on Export, select file format, share through other smartphone app, upload to third party social
media or synchronized to cloud (Figures 4-27 and 4-28,). Click on Continue & Replace if using free
account (Figure 4-29). SiteScape offers PLY and E57 format for exporting, while common mesh
model is not supported.

Figures 4-27, 4-28, and 4-29: SiteScape post scan exporting and synchronizing
4.1.2 Smartphone scanned data usage
3D point cloud data can be used in heritage conservation documentation in a variety of ways. For
example, it can be used to generate 2D screenshots from 3D point clouds, which can be used to create
detailed maps and diagrams of the heritage site. Additionally, 3D point clouds can be used to generate 3D
models of the heritage site, which can be used to create virtual tours and interactive experiences. Finally,
3D point clouds can be used to take measurements of the heritage site, such as the size and shape of
buildings, monuments, and other features. This data can be used to create detailed records of the heritage
site, which can be used to inform conservation efforts.
Experiments done in overlapping point clouds with CloudCompare can be found in Appendix B.

96

4.2 Smartphone scanning overview and directions for site survey
The app used for the data acquisition process is SiteScape. The app offers two parameters: point density
and point size. Point density determine the number of points describing the space, and point size only
impacts the visualization but not the data. Depending on the size of site and the level of detail needed,
point density can be selected from low-medium-high. In order to scan a larger area, a low level of point
density was selected.

Figure 4-30: Heritage conservation survey by smartphone workflow diagram

4.2.1 Data acquisition
Multiple scans with the same parameter were conducted with an iPhone 13 Pro. Each scans share an
overlap area for registration. In the test scans, parameters were set to high point density. The space was
digitized from right to left, including the floor and ceiling. The same scanning path and setting guarantee
the minimum difference caused by human factors in the comparison process.

97

During scan, mis-registration can be avoided through keeping more features in screen. Plain surfaces can
cause LiDAR sensor mistakes in registration and acquisition.
I. Preparation
Before conducting scans, software, registration, and parameters should be prepared.

a) Repeat 4.1.1 I. Preparation, set point density to Low and point size Low.

II. Data Acquisition
The data acquisition process remains the same as smartphone acquisition for heritage conservation
documentation. For the purpose of survey, multiple scans of abutting area are required.

a) Two scans will be conducted repeating 4.1.1 II. Data Acquisition with method A or B
b) Scan maximum area in one scan, when conducting the next scan, ensure areas of overlapping.

III. Save and Export
Post scan procedures including rename, save, and export.

a) Repeat 4.1.1 III. Save and Export.
b) Save and rename scans.

4.2.2 Data processing
Scanned data shall be downloaded to computer in PLY or E57 file format. The SiteScape webpage
(https://app.sitescape.ai/projects) can be used to show the point cloud in digital 3D space. It has functions
such as automatic registration with metric or imperial system, floorplan, measurements, and download.

98

The number of points obtained in a scan can be read at the right bottom of the interface. The built-in
measurement tool accuracy is up to 0.1 inch.

For a space that cannot be captured in single scan, scanning files should be downloaded individually as
PLY files and named systematically for data processing.

I. Download File
File can be exported from the smartphone to a computer for further data processing.

a) If the scanned data is synchronized to cloud, users can go to https://app.sitescape.ai/projects, log into
SiteScape account (Figure 4-31).

Figure 4-31: SiteScape webpage log in

b) On the website, users should see the synchronized file; click to open (Figure 4-32).

99

Figure 4-32: SiteScape project database

c) Use tools from SiteScape to measure, view floorplan, and change point size (Figures 4-33, 4-34, and
4-35).

Figure 4-33: SiteScape measurement

100

Figure 4-34: SiteScape floorplan

Figure 4-35: SiteScape change point size

d) If not synchronized to the cloud, download scanned data to a desktop computer, then opened in
CloudCompare (Figure 4-36).

101

Figure 4-36: Download options from SiteScape webpage

e) Click File-Open, change file format to PLY mesh, select the scan file, and open in CloudCompare.
When Ply File Open window pop up, click Apply all (Figure 4-37).

Figure 4-37: Open file in CloudCompare

102

II. Automatic Noise Cleaning
To reduce the fuzziness of scanned data, an automatic noise cleaning process will be completed through
adjusting following parameters:

a) Open the scan file in CloudCompare. Select the cloud in the DB Tree Window. The DB Tree
Window is the menu on the left (Figure 4-38).

Figure 4-38: Select the cloud in DB Tree window

b) Filter out floating point by clicking on the SOR icon on the tool bar and adjust the
“Standard deviation multiplier threshold” to 2.5. A new file will then be created by the software
(Figure 4-39).

103

Figure 4-39: SOR (Statistical Outlier Removal) setup

c) Select the filter file by using Tool – Other – Compute Geometric features; select “Roughness”
then adjust Local neighbor radius to 0.04. click OK (Figures 4-40 and 4-41). A scalar field will be
created. A scalar field is simply a set of values. As each value is associated to a point it is possible
to display those values as colors or to apply filters on them.

Figures 4-40 and 4-41: Compute geometric feature and settings

d) To clean up scans, go to Edit – Scalar fields – filter by value, setting the range from 0 to 0.01, and
export (Figures 4-42 and 4-43). A new file with the cleaned-up scans can be exported by clicking

104

the save icon. Color mode can be changed from Properties – Colors – RBG or Scalar filed (Figure
4-44).

Figures 4-42, 4-43, and 4-44: Filter by value, settings, and change color

III. Segment
The segment tool is used to crop point cloud data and remove unwanted points.
a) Click on the scissor icon for segment, left click to frame the data with green contour lines,
right click to finish framing (Figure 4-45).

105

Figure 4-45: CloudCompare Segment tool

b) Click on the button with a pentagon in red to remove unwanted points outside the frame. Then click
the green check button to finish (Figure 4-46).

Figure 4-46: Segment tool bar

c) In DB Tree view window, check and uncheck unwanted part to see before and after cropping. Select
the cloud of unwanted parts and delete (Figures 4-47, 4-48, 4-49).

106

Figures 4-47: Delete unwanted part

Figures 4-48 and 4-49: Before and after segmenting

d) Save the point clouds and rename the file.

Apply automatic noise clean and segment to all individual scans.

IV. Merging
The Merging tool in CloudCompare is used to piece multiple individual scans together in generating a
larger floorplan.

107

a) Open all cleaned files in CloudCompare. The scans will be registered automatically (Figures 4-50, 4-
51, and 4-52).

108

Figures 4-50, 4-51, and 4-52: Point clouds automatically registered and merged

b) Scans will automatically registered and linked. Select all file in DB Tree window and click on merge
multiple scans button on tool bar (Figure 4-53). Click yes to question window (Figure 4-54). A new
merged file will be created (Figure 4-55). Scans will be show in different color for distinguish. To
view color mode, change in Properties – Color – RGB/Scalar field (Figure 4-56).

109

Figures 4-53 and 4-54: Select all point clouds and merge

110

Figures 4-55 and 4-56: After merge view in scalar field and how to change to RGB color

c) Sometimes scans acquired from different time may be registered to the wrong place (Figures 4-57, 4-
58). In this case, users can adjust the scan with manual method or tool align two clouds by picking
four points.

111

Figures 4-57 and 4-58: Example of mis-registration of point clouds

112

d) To manually register the scans, click on translate/rotation button . A point cloud can be moved
in X, Y, and Z axis (Figure 4-59). Selected cloud will show in yellow box. Users can drag, rotate, and
move the selected scan to match the existing combination through visual (Figures 4-60, 61, 62). The
selected cloud was moved in all view direction. This method heavily relies on the operator’s ability to
edit the scan.

Figures 4-59, 4-60, 4-61, and 4-62: Translate/rotation setting, and manual alignment process

e) To register point clouds by picking points, users need to use translate/rotate function move the mis-
registered scan away from the existing scan (Figure 4-63).

113

Figure 4-63: Move to-be-aligned point cloud away from reference point cloud

f) Select both scans, then click Align two clouds by picking four points button. Select the scan you want
to align, and OK (Figures 4-64, 65, and 66).

114

Figures 4-64, 4-65, and 4-66: Align two clouds by picking four points tool, select to-be-aligned entities
window, and alignment interface

g) Pick at least three pairs of equivalent points on both clouds. Selecting points on vertical, horizontal
and different planes, avoiding all reference points on the same flat surface, can help to create a more
accurate align result (Figure 67). Click align, visual check if the scan is aligned correctly. If yes, click
green check, if not, reset and reselect equivalent points (Figure 4-68).

115

Figures 4-67 and 4-68: Pick equivalent points on both to-be-aligned and reference entities, and align
result

h) Select all files in DB Tree window. Click on merge multiple scans button on tool bar. Click yes to
question window. A new merged file will be created. Scans will be show in different color for
distinguish. To view color mode, change in Properties – Color – RGB/Scalar field (Figure 4-69)

116

Figure 4-69: Merging result in scalar field
i) Save the file.

Section View for Mapping

Section view can be created to use as floorplan or map.

a) Change to the sideview with the perspective tools by clicking on cubs on the left tool bar (Figure 4-
70).

117

Figure 4-70: Merged library scans from sideview

b) Use the segment tool to create a horizontal slice for floorplan; then move to top view to see floorplan
(Figures 4-71, 4-72).

118

Figures 4-71 and 4-72: Creating a slice of structure with segment tool, bird eye view of the floor plan

c) Users can export the floor plan file, to be traced over in AutoCAD, Revit, or other drawing software,
or measure the floor plan through uploading the file to the SiteScape webpage for measuring distance.

119

4.3 Smartphone scanning overview and directions for monitoring
Non-destructive detection of building shifting requires monitoring of a building over several years.
However, the test scan and case study were finished in a year, which does not provide enough time
change to properly study this feature. Instead of monitoring changes over time, the accuracy of point
clouds acquired by smartphone would be accessed. Different from the goal of site survey and
documentation, monitoring a building need to validate the difference of two scans that described the same
target. If smartphone captured data are stable and repeatable, they can be used for monitoring. If the
difference between the two exact same scans overly various, researchers cannot determine the origin of
deviation (from scan error/limitation or building movement).

The same physical space in the Hoose Library of Philosophy will be scanned with the same smartphone
device and SiteScape twice. The two scans will be compared in CloudCompare. In the Chapter 5 case
study, the methodology will have additional scan from a professional scanner, which will be considered
ground truth.

4.3.1 Data acquisition
Two scans in same parameter setting were acquired with the device. In the test scans, parameters were set
to high point density and medium point size. The space was digitized from right to left, including the
floor and ceiling. The same scanning path and setting guarantee the minimum difference caused by
human factors in the comparison process.

During scan, mis-registration can be avoided through keeping more features in screen. Plain surfaces can
cause LiDAR sensor mistakes in registration and acquisition.

120

I. Preparation

Before conducting scans, software, registration, and parameters should be prepared.

a) Repeat 4.1.1 I. Preparation, set point density to Low and point size Low.

II. Data Acquisition

The data acquisition process remains the same as smartphone acquisition for heritage conservation
documentation. For the purpose of survey, multiple scans of abutting area are required.

a) Two scans will be conducted repeating 4.1.1 II. Data Acquisition with method A or B
b) Scan with the same path or location twice.

III. Save and Export

Post scan procedures including rename, save, and export.

a) Repeat 4.1.1 III. Save and Export.
b) Save the two scans with the same scan path as scan 3 and scan 4.

121

4.3.2 Data processing
Scans acquired from 4.3.1 were overlapped in CloudCompare to measure accuracy. Accuracy is described
as the repeatability of scans. The cloud to cloud distance function in software was used for measure how
close the points are to each other in repetitive scans. If the distance is small enough, the methodology can
be proved valid. If the distance is overly big, the device and data cannot distinguish the change of
building over years.

Before computing the distance between the two scans, they first have to be aligned in the software. The
two scans were selected and computed with the finely registers already (roughly) aligned entities (clouds
or meshes) function with parameters. Random sampling limit and final overlap percentage can be
changed based on total number of points and the scanning difference between two scans. For example,
scan 3 compared to scan 4 has fewer points describing the ceiling, which some points lack a point to align
to. Thus the final overlap was left to 97% (Figure 4-73).

Figure 4-73: Monitoring data process work flow

122

I. Download File
Download smartphone scanned file for computer process.

a) Download scanned data in PLY file format to desktop; open CloudCompare (Figure 4-84).

Figure 4-74: Open file in CloudCompare

b) Click File-Open, change file format to PLY , select the scan file and open in CloudCompare. By this
step, the scanned data should be able to view in CloudCompare in point clouds.

II. Align Scans
Prepare two scans through CloudCompare aligning tools for distance computation.

a) Open both scan 3 and scan 4 (repeat of scan 3) in one CloudCompare window (Figures 4-75, 4-76).

123

Figure 4-75: Scan 3 point cloud in CloudCompare

124

Figure 4-76: Scan 4 point cloud in CloudCompare

b) Select both scans in DB Tree window, click finely registers already (roughly) aligned entities (clouds
or meshes) button; then a clouds registration window will appear (Figure 4-77).

Figure 4-77: Finely registers already(roughly) aligned entities (clouds or meshes) tool

c) Click on parameters on clouds register window, adjust final overlap to 97%, and select adjust scale
(Figure 4-78).

125

Figure 4-78: Cloud registration window setting

d) Click on research on clouds register window, adjust random sampling limit to 50,000, rotation to
XYZ, and select all for translation, then click OK (Figure 4-79). The final Root Mean Square smaller
than 0.5 represents a good result.

126

Figure 4-79: Result of aligning scan 3 and scan 4

III. Distance Computing
The two-point clouds were compared respectively with following steps.

a) Select both scans in DB Tree window.
b) Click “Compute Cloud to Cloud Distance” icon on top tool bar, here scan 4 is set as reference and
Scan 3 as comparison (Figure 4-80).

127

Figure 4-80, Scan 4 as Reference and scan 3 as Compared in CloudCompare cloud to cloud distance
function

c) On the pop up window General parameter, select AUTO on Octree level, multi-threaded max thread
count 8/8, Local modeling – None (Figure 4-81).

128

Figure 4-81: Distance computation parameter setup

d) On approximate distance window, click on the bottom right bar chart histogram, the chart will show
the maximum distance and roughly estimated distance (Figure 4-82).

Figure 4-82, Approximate distance and Histogram for scan 3 and scan 4.

e) Click Complete then the Cloud to Cloud distance will be computed.

129

f) In the Properties window, change saturation to further analysis the range and portion of distant
points.

The distance computation window provided an approximate distance between the two scans, and the
histogram provided a graphical exhibition of the range of distance with a maximum distance of 1.080
meters and an average distance of 0.009 meters.

After the computation was completed, Figure 4-83 would be available from CloudCompare visualizing
cloud-to-cloud distance between Scan 3 and Scan 4. Since Scan 4 was the reference, it was shown in
RGB color; Scan 3 as the compared scan was shown in the scalar field. Select the Scan 3 (compared scan)
from DB Tree window and scrolled down the Properties window, a Scalar Field (SF) displayed
parameters could be seen and adjusted.

The default setting after distance computation has a saturation ranging from 0.00003713 meters to
1.08476901 meters (Figure 4-84). White dots can be moved around to change the color range of
saturation. Displayed option determines the range of saturation, points with a distance bigger than the
setpoint displayed will turn grey. Saturation helps visually describe the distance between two clouds. The
difference in color exhibit point distance in corresponding point clouds.

Figure 4-83, Cloud-to-cloud distance visualization from CloudCompare for scan 3 and scan 4.

130

Figure 4-84, Saturation display range.
Adjust the saturation to 0.0254 meters (1 inch) and kept displayed at 1.08 meters; the Cloud to cloud
distance and saturation display range changes (Figure 4-85). Changes in saturation helped represent the
distance between clouds. Red over green over blue represented the distance from large to small. With a
saturation setpoint of 0.0254 meters, red and green can be observed on corners, decorative panels, ceiling,
and furniture, which represented a deviation of these areas on Scan 3 compared to Scan 4.

Figure 4-85: Scan 3 and Scan 4 distance with saturation 1 inch. Distance reduced from red to green to
blue.

The results showed that there were areas such as bookshelves, statues, floors, and major building
structures of both scans overlapping each other with a distance of less than 1 inch. However, ceilings,
sofa, chair, and some floor areas were deviated larger than 1 inch, representing in red.

While with the current methodology, the scans still might not be useful solutions for heritage preservation
users, who want to monitor the changes of building overtime, because the researcher was unable to

131

validate the repeatability of scans and captured points. However, the result does not negate the potential
of the methodology because the following possible impacting factors cannot be ruled out 1) the data
acquisition path for both scans were not exact same. Different scanning routes may cause the deviation, 2)
scan alignment did not exact overlap the two scans, 3) the distance computation in CloudCompare only
measured nearest neighbour distance, but not the true distance (Figure 4-86). Thus, the methodology for
monitoring heritage conservation purposes is not recommended until further research been done in the
control of scanning path, the validation of alignment computation, and the measurement of the true
distance between two point clouds. The results only exhibit a preliminary assessment to smartphone
acquired data, further comparison will be made in Chapter 5.

Figure 4-86: CloudCompare distance computation working principle

4.4 Smartphone scanning overview and directions for Education
Matterport Capture smartphone application was used in creating virtual tour for heritage conservation
education exhibition purposes. A data acquisition and data process instruction will be given. A tripod is
recommended for following the start-up guide.

132

4.4.1 Data acquisition
Data acquisition was completed with the Matterport Capture app on the smartphone device with a tripod.
Different from the SiteScape app which allows free movement during capture, Matterport Capture
required the device to be stable in one place. A tripod was used helping fixed the device at the same time
able to turn in spherical degrees.

I. Preparation
To generate a virtual tour through smartphone, Matterport Capture should be downloaded, registered, and
set up.

a) Have smartphone application Matterport Capture download to the device, and smartphone stabled on
a tripod.

b) Open Matterport Capture and sign up a free account (Figure 4-97, 4-98).

133

Figure 4-87 and 4-88: Matterport Capture register interface

c) Remove unwanted furniture or objects from the physical space.

II. Data Acquisition
Spatial information can be acquired through Matterport Capture and therefore create a virtual tour.

a) Go to My Jobs, click on + New Job. Fill in address information (Figure 4-89).

134

Figure 4-89: New job address information

b) Click Option, select 3D Scan and Complete Scan (Figures 4-90, 4-91).

Figures 4-90 and 4-91: Scan interface and parameter selection

c) Follow the instructions on the screen to point the camera at the dots, move to next one until finished
(Figures 4-102, 4-103)

135

Figures 4-92 and 4-93: Scan process

d) Take at least three to four 360 degree photos in one space, more when documenting narrow space and
doorways.

e) After a scan is finished, trim and add window to the model on the smartphone app (Figure 4-94).

136

Figure 4-94: Trim and add window interface

f) Upload to cloud one at a time.

g) For each room, include at least three captures from different standing points. Ensure overlap area
between two scans, which enhance the automatic registration.

4.4.2 Data processing
Matterport captured 360 photographs can be uploaded, edited, and shared publicly.

I. Data Editing
Data editing can be done with Matterport project website, which includes various built-in functions such
as measurements and different views.

137

a) Go to Matterport webpage: https://matterport.com. Click on Get Started Free, log in Matterport
Capture account (Figures 4-95, 4-96).

Figure 4-95: Matterport webpage

138

Figure 4-96: Matterport sign in

b) Once the project shows uploaded, users can view their projects on the website (Figures 4-97, 4-98).

Figures 4-97 and 4-98: Matterport project uploaded, and Matterport Space

c) Open the project on the web page; click on start button to explore the building in virtual space (Figure
4-99 ）.

139

Figure 4-99: Matterport project space

d) Move around by clicking on white circles in view to check different views (Figures 4-100, 4-101).

140

Figures 4-100 and 4-101: white circles and different views of a project

e) Matterport enables a doll house view, floorplan, and measurements (Figures 4-102, 4-103, 4-104, 4-
105)

141

Figure 4-102: Doll house view

Figure 4-103: Floorplan view

142

Figures 4-104 and 4-105: Measurements can be taken from any view mode

II. Sharing

143

To create an accessible virtual tour and share, researcher should adjust details and privacy settings on
Matterport project webpage.

a) Click on Details, adjust information (Figure 4-106)

Figure 4-106: Matterport project details

b) Additional services such as export to an E57 file (If data is scanned with LiDAR mode, it can be
exported as an E57 file as point clouds), a BIM file, and schematic plan can be all be ordered from
Matterport for a charge (Figure 4-107). Various add-ons offered by Matterport allow users to further
extract information from the scan for usages in different fields.

144

Figure 4-107: Matterport add on features

c) Click on Share and select private for personal use, password protected for personal share or public for
general public. Users can select level of detail and information they want to share through options
(Figure 4-108).

145

Figure 4-108: Share and Invite page

d) Copy link to share or insert into individual project websites. Matterport can also be embedded in
users’ own website through using embed code, for more detailed instruction see
https://answers.pagecloud.com/help/matterport and https://support.matterport.com/s/article/Embed-a-
Matterport-3D-Model?language=en_US. Anyone with a shared, public accessed link and internet can
view the virtual tour. If the virtual tour is embedded in a website, it can be view from the webpage.

4.5 Summary
Chapter 4 is mainly a user guidance for users interested in gathering 3d point cloud data. It gave detailed
instruction on how to scan and use the scanned data for different purposes.

146

• 4.1 introduced the process taking multiple scans of an interior with SiteScape and merging them
together to generate floor plan. The test scan exhibits the possibility of creating the product of a
floorplan, which will be conducted again in the case study and compared with data acquired from
a professional scanner.

• 4.2 described scanning the same area with point density parameter in low, medium, and high with
SiteScape. Two of the three scans were processed and used for comparing the possibility of
overlaying scans to increase accuracy. The validity of this methodology will be examined in case
study through comparing with scanner’s data as ground truth in Chapter 5.

• 4.3 introduced the process of acquiring geometric data of the same area with same parameter in
SiteScape. CloudCompare was used to compute the distance between the two exact scans (scan 3
and scan 4) to examine the repeatability of smartphone scanned data. 4.4 used Matterport Capture
for iPhone and a tripod to take 360 degree photographs and generate a virtual tour. The file was
uploaded to website and can be shared through links.

• 4.4 described how to use create a smartphone 3d virtual tour with Matterport Capture that can be
shared through links on a website.

The results from test scans show that all four methodologies are relatively straightforward. For site
survey purposes, a floor plan was created. Further comparison between the floor plan to scanner
scanned data will help in examine the hypothesis. Using smartphone 3d scan to document cultural
heritage was shown; the result shows overlapping two scans can increase the points describing the
same surface,; however, a thicker layer of point may not contribute to documentation because of
increased error. In test scans for monitoring purposes, two scans with the same parameter and a scan

147

path was used to computed the cloud to cloud distance. The distance exhibited that smartphone
scanned data cannot produce repetitive data every time unless with more stable scanning methods or
better alignment computation. Issues found in current methodology for documentation and
monitoring will be discussed further in Chapter 5 Case Study

148

Chapter 5 Case Study Smartphone Scan Result
Chapter 5 discusses the results of 3d scanning from an iPhone and a professional scanner and a laser
measure at the case study Reunion House. The Reunion House is located at Silverlake, California. The
project was built in 1951, designed by Richard and Dion Neutra. The case study takes the master bedroom
as an example, scanned by both smartphone and professional scanner device (Figure 5-1). Christopher
Gray, Associate Reality Capture at GB Geotechnics USA Inc (GBG), introduced Alan D. White to
conduct the professional scan. White is a technician of AQYER, a company specializing in non-
destructive evaluation and as-built documentation, has made a significant contribution to the research
project. He conducted a professional scan for the sake of Neutra Reunion House: Documentation
Workshop lead by Western Chapter of the Association of Preservation Technology International and
generously shared the scan data (WCAPT, 2023). The professional scanner used is a Leica RTC 360
scanner, which is commonly used for small and medium project. White’s data has enabled the comparison
between the iPhone scan and professional scanner scan. His commitment to the preservation of the Neutra
Reunion House is unparalleled, and his contribution to the project is highly appreciated.

Figure 5-1: Master bedroom in Reunion House

149

The measurement results from smartphone scans, professional scanner scans, and laser distance
measurements data will be summarized. Smartphone scanned results will be compared to the other three
methods. Scans acquired from smartphone and professional scanners will be imported to AutoCAD.
Dimensions of all three methods will be compared and a range of error will be calculated. The
comparison results will be summarized in a table. Recommendations will also be made based on the
comparison results. The results will help to guide the selection of the most appropriate method for a
specific application in Chapter 6.

5.1 Evaluation standard
When generating drawings and models for heritage conservation, it is important to ensure that the scale
and standard of accuracy are appropriate. The smartphone scan was evaluated based on HABS
requirements, which accepts paper files. Recording Historic Structures introduced the scales used for
architectural drawing (Table 5-1) (Burns, 2004). Different from digital drawings, drawing on paper scale
is related to the size of paper. With fixed size of board and paper, scale and line weights can impact
accuracy. The most common architectural scale is 1/4” = 1’-0”, with the smallest unit of 1” (Burns, 2004).
Such scale includes reasonable amount of detail possible. To document windows and doors, and other
features of similar scale, 3/4”= 1’-0” scale with a smallest unit of 3/8” are mostly used. To test the ability
of smartphone scanned data in assisting as-built drawings on paper, scale 1/4” = 1’-0” and 3/4” = 1’-0”
would be used in evaluating the performance of scanned data. To identify the smallest unit for scale
1/4”=1’-0” and 3/4”=1’-0”, accuracy/ precision should reach or smaller than a factor of two than the
smallest feature, thus 1/2” or 3/16” (Burns, 2004). Digital documentation always require a 1/8” of
accuracy for tracing room size three dimensional models, floor plan, or section drawings (White and
Gray, 2022). To create engineering drawing and maps, the most common scale is 1”=20’ with a smallest

150

unit 0.4’ (Burns, 2004). In order to determine 0.4’ in a map, a factor of two than the smallest feature,
which is 0.2’ (2 2/5”) is needed.
When evaluating smartphone scans, the smallest units of 1/2” or 3/16”, 1/8”, and 0.2’ (2 2/5”) will be
used as a standard (marked in red).
Table 5-1: Architectural Scales for drawing (Burns, 2004)
Scale
Smallest
Unit
Evaluating
Unit
Use
1/16" = 1'-0" 4" 2"
Drawings of large structures without
details.Materials shown in plan only.
1/8"=1'-0" 2" 1"
Little detail possible. Materials shown in
plan, only large units in elevation.
1/4"=1'-0" 1" 1/2"
The most common architectural scale.
Reasonable amount of detail possible.
HABS/HAER shows door and window
frames, materials in both plan and
elevation. At this scale, line weights can
adversely affect accuracy. A 3x0 (0.25
mm) line is approximately 1/2" thick.
3/4"=1'-0" 3/8" 3/16"
Most common scale for door/window
elevations and other features of similar
scale.
1 1/2"=1'-0" 3/16" 3/32"
Details of door/window jambs/frames,
large tools, small machines, etc.
3"=1'-0" 3/32" 3/64"
Details of objects such as hardware, tools,
etc. and molding profiles.
Full Size
Small or intricate objects, elaborate
moldings and ornamentation

151

Table 5-2: Engineering and map scales (Burns, 2004)
Scale
Smallest
Unit
Evaluating
Unit
Use
1"=5,280' 104' 52' USGS 15 minute map
1"=2,000' 40' 20' USGS 7.5 minute map
1"=40' 0.8' 0.4' Site map
1"=20' 0.4' 0.2'
Very common scale for residential-size
site plans (at this scale a half-acre lot fits
comfortably on a legal-size page in a deed
book). Distances given in feet and
hundredths.
1"=16.66' 0.33' 0.165' Site map
1"=10' 0.2' 0.1' Small site map
1"=8.33" 0.166' 0.083' small site map

5.2 Scanned data results
Scans were exported as point clouds in E57 file. After being processed in Autodesk Recap, the file was
saved in rcp. format then imported to AutoCAD. In AutoCAD, RTC 360 scan was set as the ground truth;
the iPhone scanned point clouds would be aligned to it. Then, measurements were documented and
compared. To control impact factors, six targets are set in the comparison: 1) master bedroom height 2)
master bedroom width 3) master bedroom length, 4) master bedroom closet to closet distance 5) desk
height, and 6) point clouds thickness (Figure 5-2). The six measurements were documented and listed in
tables for comparison.

152

Figure 5-2: Reunion House master bedroom and six targets

Seven smartphone scanned point clouds were selected for the comparison, including
1) a raw low point density scan on tripod,
2) a raw high point density scan,
○
2
○
3
○

153

3) a merged high point density scan,
4) a raw low point density scan,
5) a merged low point density scan,
6) a raw floor to ceiling scan, and
7) a merged floor plan from multiple scans.
Smartphone scanned data acquired on tripod was compared to scanner scanned data. Professional scanner
laser sensor rotated 360 degree in horizontal and vertical axis except for a small area where the tripod
stood (Figures 5-3 and 5-4). Smartphone device was stabled on tripod; during a scan, smartphone only
rotate on vertical axis (Figures 5-5, and 5-6). Limited by the angle of smartphone LiDAR sensor, ceiling
and floor had two circles of blank area. Impacted by the angle of laser shooting on surface, ceiling and
floor are barely captured. Tilting smartphone up and scan can create less blind area on ceiling; tilting
smartphone down can create less blind area on floor (Figures 5-7, 5-8, 5-9, and 5-10). A more complete
scan can be merged with two angled scan (up and down) and a leveled scan. (Figure 5-11).

Figures 5-3 and 5-4: Professional scanner scan and moving path

154

Figure 5-5 and 5-6: Smartphone scan moving path and scan result

Figures 5-7 and 5-8: Smartphone device tilted up and scan result

155

Figures 5-9 and 5-10: Smartphone device tilted down and scan result.

Figure 5-11: Three smartphone scans (leveled, upper and lower angled) merged with CloudCompare

Smartphone scans (red) were overlapped and aligned manually with professional scanner captured point
clouds (green) (Figure 5-12). Three floor plan sections were created to observe the difference and
measure the distance between smartphone scans to Leica RTC 360 scanned data in Autodesk (Figures 5-
13, 5-14, 5-15, 5-16). The measurements were done through tracing the point cloud and using the linear
dimension tool. Dimensions were documented in tables; error and error rate were calculated. Results were
analyzed and compared to determine which scan yielded the most accurate measurements in 5.2.1.

156

Figure 5-12: Manual align smartphone scans (red) with professional scans (green).

Figure 5-13: Section layout on point clouds

157

Figure 5-14: Section 1 in AutoCAD

Figure 5-15: Section 2 in AutoCAD

158

Figure 5-16: Section 3 in AutoCAD

A section drawing is a vertical cut through a structure or site, which provides vertical information and
reveals the arrangement of objects and spaces. It shows a series of room elevations, separated by walls,
floors, and ceilings, in relation to one another. Sections are similar to floor plans except they are cut
perpendicular to the floor, and the visible surface beyond the cut line are room elevations instead of a
floor. Tracing and measurements from a section can only include room height, as room length and width
cannot be determined in the same section.

Room dimension is measured individually in all eight scans through the plans and sections. In each
measurement, the section was first traced and then measured (Figure 5-17). Because laser only travels in
straight line, scans can have blind areas. The blind areas in iPhone acquired scans that are not applicable
for measurements are shown N/A in tables. One wall was missing in smartphone scanned data; thus the
width of room cannot be measured (Figures 5-18 and 5-19).

159

Figure 5-17: Measurement example (room width) of iPhone acquired low point density merge scan (red)
and professional scan (green) in section 1

160

Figures 5-18 and 5-19: Blind area (outlined in blue) observed from overlapped floorplan and from section
2 (high point density scan by smartphone in red).

Tracing of point clouds is a decision-making process. To measure the dimension of room, the most
interior points are traced and measured. Taking the master bedroom height measurement as an example,
the highest point describing the floor and the lowest point describing the roof were traced and measured
(Figures 5-20, 5-21, 5-22).
In the case where ceiling, floor, wall, or the surface being capture is not straight, the measurement
location should be the same to reduce human factor impact.

161

Figure 5-20: Master bedroom floor to ceiling measurement example

Figure 5-21: Master bedroom height measurement ceiling tracing detail. Lines traced the top of ceiling.

162

Figure 5-22: Master bedroom height measurement floor tracing detail. Lines traced the bottom of the
floor.

The thickness of points in point clouds can impact tracing and measuring result. RTC 360 scanned point
cloud shows a consistent thin layer of points, the top and bottom points are relatively easy to determine
and trace (Figure 5-23). iPhone scanned data varies based on different point density. High point density
mode results in a thicker layer of points (1/2 inch – 1 1/4 inch), low point density mode results in a
thinner layer of points (1/4 inch – 1/2 inch). The thickness and fuzziness of points were impacted by
distance and captured angle. The closer and more perpendicular to scanning target, the points will be
thinner and organized. High density iPhone scans were fuzzier and denser, when zoom in it is easier to
observe points. Low point density iPhone scan points were less concentrated with less floating points.
Comparing to the consistence 1/8” thickness of RTC 360 scanned point cloud, iPhone scanned point
clouds’ thickness are more irregular; in one scan, thickness can range from 1/4 inch to 1 1/4” inch. The
fuzzy character of iPhone scanned point cloud create a difficulty in tracing because the majority of points
are located in a densely described area, but a few points are flying around (Figure 5-24). There is no way
to determine if the flying points are mis-captured fuzziness or an accurate capture. In this case, even the
fuzzy points are questionable, they are still taking for tracing and measurement since the credibility is
undistinguishable.

163

Figure 5-23: Tracing top and bottom points in section 2 with line command

Figure 5-24: Measuring point thickness with dimensional tool. RTC 360 point cloud thickness is 1/8”,
iPhone scanned point cloud thickness is 1”.

164

5.2.1 Smartphone scanned data results
The seven smartphone scans were processed through Autodesk Recap, which allowed for the data to be
exported to rcp. files. These files were then imported into to AutoCAD, where all seven scans captured by
the iPhone 13 Pro were measured using AutoCAD’s dimensional tools.

Dimensions of room height, width, closet to closet distance, desk height and point cloud thickness were
then compiled into tables (Tables 5-3 to 5-7). The measurements from table were used in comparison to
measurements from the professional scanner and laser measurements to determine the accuracy of the
scans and to assess the overall quality of the data.

Table 5-3: iPhone scanned high density point clouds measurements
iPhone High
Density
Master
Bedroom height
Master
Bedroom Width
Master Bedroom Closet
to Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A N/A N/A N/A N/A
Section 2 7'-9 1/2" N/A N/A N/A 1 1/4“
Section 3 7'-9 3/4" N/A N/A 2’-5” 2 1/2”

Table 5-4: Merged high density scans by iPhone measurements
iPhone High
Merge
Master
Bedroom height
Master
Bedroom Width
Master Bedroom Closet
to Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A 15'-7 3/8" N/A N/A N/A
Section 2 7'-8 5/8" N/A 10’-10 5/8” N/A 5/8”
Section 3 7'-9 3/4" N/A N/A 2'-5" 3/4”

Table 5-5: iPhone scanned low density point clouds measurements
iPhone Low
Density
Master
Bedroom height
Master
Bedroom Width
Master Bedroom Closet
to Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A 15'-10 1/4" N/A N/A N/A
Section 2 7'-11 1/4" N/A 10’-8” N/A 1 1/8”
Section 3 7'-9 5/8" N/A N/A 2'-5 1/2" 5/8"

165

Table 5-6: Merged low density scans by iPhone measurements
iPhone Low
Merge
Master
Bedroom height
Master
Bedroom Width
Master Bedroom Closet
to Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A 15'-8 1/8" N/A N/A N/A
Section 2 7'-11 1/4" N/A 10'-8 7/8" N/A 1 1/2“
Section 3 7'-11 1/4" N/A N/A 2'-4 5/8" 1 5/8"

Table 5-7: Merged floorplan measurements
iPhone Merged
Floorplan
Master
Bedroom
height
Master
Bedroom
Width
Master Bedroom Closet
to Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A 15‘-9 5/8 N/A N/A N/A
Section 2 7'-11 3/8" N/A 10'-8 7/8" N/A 1 1/4”
Section 3 7'-11 1/4" N/A N/A 2'-5 1/8" 3/4"

5.2.2 Professional scanner scanned data results
A 3D laser scan was done with a Leica RTC 360 LiDAR scanner by Alan D. White. Three scans were
captured with the 3D scanner set on a tripod at three different locations in the master bedroom. The point
clouds were registered in Leica Cyclone software and translated to an RPC file in Recap. Measurements
were then taken using AutoCAD tools (Table 5-8). The data obtained from this professional scanner was
used as a ground truth for comparison with data obtained from a smartphone scanner. This comparison
will help to evaluate the accuracy of the smartphone scanner.

Table 5-8: Leica RTC 360 scan measurements
Leica RTC
360
Master Bedroom
height
Master Bedroom
Width
Master Bedroom Closet to
Closet Distance
Desk
Height
Point Clouds
Thickness
Section 1 N/A 15’-9 1/8” N/A N/A N/A
Section 2 7’-10 1/2” N/A 10’-8” N/A 1/8”
Section 3 7’-10 1/2” N/A N/A 2’-5” 1/8”

The scanner was set to low density mode during capture; a thin layer of points were exhibited in the point
clouds (Figure 5-25) The thickness of data is 1/8” and remained stable in the whole point cloud.

166

Figure 5-25: Layer of points from Leica RTC 360

5.3 Result comparison between iPhone and Leica RTC 360
Dimensional data exhibited in previous table (Table 5-1 – 5-6) were reorganized for comparison and
calculating error. Observation in visual comparison between RTC 360 point cloud and iPhone scanned
point clouds were shown through screenshots.

5.3.1 Measurements and errors
The measurement results from iPhone 13 Pro were compared to the data acquired from professional
scanner RTC 360 individually and compared (Tables 5-9 to Table 5-14). Point cloud acquired by Leica
RTC 360 was imported to AutoCAD and used as the ground truth values.

167

Table 5-9: Room width measurements comparison in section 1
Section 1
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Room Width
15'-9
3/4" N/A 15'-7 3/8" 15'-10 1/4" 15'-8 1/8" 15‘-9 5/8
Error

N/A 1 7/8" 1/2" 1 5/8" 1/2”
1/2"

N/A x √ x √
3/16"

N/A x x x x
2 2/5"

N/A √ √ √ √
Error
Percentage

N/A 0.94% 0.26% 0.86% 0.26%

The percentage of error was calculated through dividing difference between iPhone scanned point clouds
and RTC 360 scanned clouds by the room width measured from RTC 360 scanned data. The highest
percentage of error happened with merged high point density scan in 0.94% within a range of 15 foot.
The lowest percentage of error occurred in low point density scan in 0.36%.

Table 5-10: Room height measurements comparison in section 2
Section 2
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Room Height
7'-10
1/2" 7'-9 1/2" 7'-8 5/8" 7'-11 1/4" 7'-11 1/4" 7'-11 3/8"
Error

1” 1 7/8” 3/4” 3/4” 7/8"
1/2"

x x x x x
3/16"

x x x x x
2 2/5"

x x √ √ √
Error
Percentage

1.06% 1.98% 0.79% 0.79% 0.93%

The percentage of error of room height measurement was calculated in section 2. The result shows
smartphone scan in low point density without process have the closest measurement to the scanner scan.
The largest error percentage was shown in iPhone scanned data in high density, then merged. High

168

density scan demonstrates a higher error rate than low density scan, which represent a larger deviation
from scanner data. Even though iPhone high merge, iPhone low merge, and iPhone floorplan were all
merged scans, there are differences in the percentage of error. The result shows error in iPhone low
density < iPhone low merge < error in iPhone floorplan < error in iPhone high density < iPhone high
merge.

Table 5-11: Bedroom width 2 measurements comparison in section 2
Section 2
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Bedroom
Width 2
10'-8" N/A 10’-10 5/8” 10’-8” 10'-8 7/8" 10'-8 7/8"
Error

N/A 2 5/8" 0 7/8” 7/8”
1/2"

N/A x √ x x
3/16"

N/A x √ x x
2 2/5"

N/A x √ √ √
Error
Percentage

N/A 2.05% 0 0.68% 0.68%

Error and error rate in bedroom width 2 measures are shown in table (Table 5-7). Low point density
iPhone scanned data exhibited a same measurement with the Leica RTC 360 scan, achieving a 100%
accuracy for a 10 foot distance. The merged floor plan scan from iPhone has a 0.68% of error, and the
merged low point density scan from smartphone has a 0.68% of error. The largest deviation was shown in
the merged high point density scan from iPhone with a 2.05% of error from professional scanner data.

169

Table 5-12: Room height measurements comparison in section 3
Section 3
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Floor to
Ceiling
Room Height
7'-10
1/2" 7'-9 3/4" 7'-9 3/4" 7'-9 5/8" 7'-11 1/4" 7'-11 1/4" 7'-11"
Error

3/4" 3/4" 7/8" 3/4" 3/4" 1/2"
1/2"

x x x x x x
3/16"

x x x x x x
2 2/5"

√ x x x x √
Error
Percentage

0.80% 0.80% 0.93% 0.80% 0.80% 0.53%

In the measurement of room height from section 3, high density scan and floor to ceiling scan from
iPhone shows the smallest error in 1/2”, resulting in 0.53% of error in 7’-10 1/2”. The largest deviation is
exhibited in merged low density scan from smartphone, causing 0.93% of error rate.

Table 5-13: Desk height measurements comparison in Section 3
Section 3
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Floor to
Ceiling
Desk Height 2'-5" 2'-5" 2'-5" 2'-5 1/2" 2'-4 5/8" 2'-5 1/8" 2'-5 1/4"
Error

0 0 1/2" 3/8" 1/8" 1/4"
1/2"

√ √ √ √ √ √
3/16"

√ √ x x √ x
2 2/5"

√ √ √ √ √ √
Error
Percentage

0 0 1.7% 1.3% 0.43% 0.86%

The error and error rates of desk height from section 3 was measured and calculated. High density scan
and merged high density scan reach 0 inch in error (with an accuracy of 1/8”). The merged floorplan
shows a 0.4% of error with 1/8” of deviation, floor to ceiling scan achieved 0.9% of error with 1/4” of

170

deviation. The performance of low density scan and merged low density scan demonstrated 1.7% and
1.3% of error, which represent more deviation from the scanner data.

Table 5-14: Point cloud thickness comparison in section 2 and 3
Point Clouds
Thickness
RTC
360
iPhone High
Density
iPhone High
Merge
iPhone Low
Density
iPhone Low
Merge
iPhone
Floorplan
Section 2 1/8" 1 1/4“ 5/8” 1 1/8” 1 1/2“ 1 1/4”
Section 3 1/8" 2 1/2” 3/4” 5/8" 1 5/8" 3/4"

Comparison in point cloud thickness were made in section 2 and 3. Points from RTC 360 scanner show a
consistent pattern and a stable thickness of 1/8” The performance of smartphone acquired points are more
fuzzy and more irregular (Figure 5-26).

171

Figure 5-26: Ceiling detail in low point density scan section 2. Green representing RTC 360 scanner data,
red representing iPhone scanned data. The thickness of scanner points is 1/8”; the thickness of iPhone
scanned points is 1 5/8”

The result from the comparisons reveals a pattern of low point density scans achieving better results than
high point density scans, and raw data performing better than merged files. The high point density scan
error rated from 0 to 1.06%, high point density merged scan error rated from 0 to 2.05%, low point
density scan error rated from 0 to 1.72%, low point density merged scan error rated from 0.68% to 1.29%.
Scanned data with original low point density mode has the best performance with the least error rate and
stabled measurements. Qualification of each mode of smartphone scans for heritage conservation
purposes and scales will be discussed in 5.3.2.
5.3.2 Visual comparison
Point clouds captured from iPhone with various parameters and set ups were imported and aligned to the
ground truth. Three section views (plans) were created to visualize the deviations and differences between
professional scanner scanned data and iPhone scanned data.

172

Scanner and iPhone acquired point cloud thickness of bedroom ceiling are visualized (Figure 5-27). In the
case of Reunion House, both scanner and iPhone captured the slope of the roof. The professional scan
described the surface with one layer of clean points. The iPhone captured geometry is fuzzy and hard to
recognize. As section 5.2 introduced, the fuzziness of points obtained by smartphone cause difficulties in
determine the accurate location of surfaces. Tracing the lowest point of ceiling for measurements can be a
possible reason for errors and impacting factor of accuracy.

Figure 5-27: High density iPhone scan (red) and RTC 360 scan (green) ceiling detail

The detail of a low-density iPhone scan is shown aligned with the point cloud from the professional
scanner in Section 2 (Figure 5-28). The detail section view described the thickness of a sliding closet
door. The green point cloud (RTC 360) displays two surfaces, and the thickness of the red point cloud is
the distance between them. The red point cloud is accurately picturing the structure and thickness,
however, without the aid of the professional scanner data, this information would be difficult to interpret.
It requires heritage conservation professionals having access to site when reading point clouds, or heritage
conservation professionals being very familiar to the site, to read and understand such detail.

173

Figure 5-28: Low density iPhone scan (red) and RTC 360 scan (green) closet detail

The iPhone scan exhibited a deficiency in describing details. The lighting structure on the right wall from
section 1 was described differently in professional scanner point cloud and the iPhone point cloud. The
structure of lighting was described clearly from professional scanner, while the iPhone captured a rough
shape at the accurate location (Figures 5-29 and 5-30). Even though not able to capture detailed geometry,
the observation supported smartphone device’s ability to describe feature location, which can be used in
assisting heritage conservation tasks.

Figures 5-29 and 5-30: Low density iPhone scan (red) and RTC 360 scan (green) section 1 (left), and
detail (right).

174

From the section views, it was obvious that smartphone was struggled to capture corners. In the example
from merged scan of high point density scan, the roof to wall corner were described as a round turn
instead of a sharp ninety degree angle (Figure 5-31). In comparison, the corner of drawer meeting ground
was described shapely. This might be impacted by distance factors that in a merged scan, three scans from
lower angle, level, and upper angle are piecing together. The scanning device is closer to ground than to
ceiling; smartphone camera was also facing ground target more perpendicular than facing roof.

Figure 5-31: Merge of high point density scans from smartphone aligned with professional scanner point
clouds

5.4 Qualification
Based on the deviations measured from smartphone scans and professional scanner scans, low point
density scan by smartphone achieved acceptable accuracy/precision in desk height, and bedroom width 2
distance for architectural drawing scale 4/1”=1’-0”. Merged floorplan scan achieved acceptable
accuracy/precision in desk height and room width measurements. Floor to ceiling scan achieved 1/4” in

175

desk height. None of the room heights were measured accurately enough for creating an architectural
drawing. With some target measurement laying outside the acceptable range, low point density scan
acquired from iPhone achieved error rates from 0 to 1.72% (Table 5-15). Bedroom width 2 distance lays
in an acceptable range of error as an alternative digital documentation method. Desk height, room heights,
and width measurement are not precise enough as a laser scan with the current method.

Table 5-15: Smartphone scan low density mode error rate
Low Density
Desk
Height
Room
Height
Section 3
Room
Width 2
Room
Height
Section 2
Room
Width
Error 1/2" 7/8" 0 3/4” 1/2"
1/2" √ x √ x √
3/16" x x √ x x
Error
Rate
1.72% 0.93% 0.00% 0.79% 0.26%

Merged low density scans achieved error rates from 0.68% to 1.29% (Table 5-16). The least deviation
happened with the measurement of desk height, meeting architectural scale 1/4” = 1’-0” smallest unit.
The most deviation occurred in room height. Scan measurements except for desk height do not meet
drawing standard. None of the measurement meet laser scan standard.

176

Table 5-16: Smartphone scan low density mode merged error rate
Low Density Merge
Desk
Height
Room
Height
Section 3
Room
Width 2
Room
Height
Section 2
Room
Width
Error 3/8" 3/4” 7/8” 3/4” 1 5/8"
1/2" √ x x x x
3/16" x x x x x
Error
Rate
1.29% 0.80% 0.68% 0.79% 0.86%

High density scans achieved error rates from 0.0% to 1.06% (Table 5-17). Desk height reaches the
standard for common architectural scale, but desk height and room height do not meet laser scan accuracy
standard.
Table 5-17: Smartphone scan high density mode error rate
High Density
Desk
Height
Room
Height
Section 3
Room
Width 2
Room
Height
Section 2
Room
Width
Error 0 3/4" N/A 1” N/A
1/2" √ x N/A x N/A
3/16" √ x N/A x N/A
Error
Rate
0 0.80% N/A 1.06% N/A

177

Merged high density scans achieved error rates from 0 to 2.05% (Table 5-18). Same as low point density
mode, desk height appeared meeting common architectural drawing scale, but none of the measurements
can be used to determine the smallest unit for architectural scale 1/4”=1’-0”.
Table 5-18: Smartphone scan high density mode merged error rate
High Density Merge
Desk
Height
Room
Height
Section 3
Room
Width 2
Room
Height
Section 2
Room
Width
Error 0 3/4" 2 5/8" 1 7/8” 1 7/8"
1/2" √ x x x x
3/16" √ x x x x
Error
Rate
0 0.80% 2.05% 1.98% 0.94%

Merged scan floorplan achieved error rates from 0.0% to 2.05% (Table 5-19). Merged floorplan scans
have the least deviation with desk height, then room width. Both measurements meet common
architectural drawing scale; measuring feature such as desk height can even be used for door/window
drawing. Room height and closet to closet distance, however, do not meet common drawing scale
smallest unit identify standard and digital documentation scale.

Table 5-19: Smartphone scans merged floor plan error rate
Floorplan

Desk Height Room Height Section 3 Room Width 2 Room Height Section 2 Room Width
Error
1/8"
3/4"
7/8” 7/8" 1/2”
1/2" √ x x x √
3/16" √ x x x x
Error Rate
0.43
0.80%
0.68% 0.93% 0.26%

178

The measurement of desk height for the five methods (low point density, merged low point density, high
point density, merged high point density, and merged floorplan) can be used to identify the smallest unit
for common architectural drawing scale, merged, high point density, and floor plan measurements of desk
height even can be used to determine the smallest unit for scale used on window and door drawings.
Thus, smartphone scans with tripod are approved for the documentation, survey, and monitoring of
furniture size architectural features.

In the measurement of room dimension, smartphone scans have limited ability. With raw scan data, low
point density is closer to the ground truth; the merged high point density scan has the most deviation from
the ground truth. None of smartphone scans measurements of room height, width, or closet to closet
distance reached professional scanner’s accuracy of 1/8”. Therefore, with current methodology
smartphone scan is not a recommend substitute for professional scanner scan in creating architectural
drawings.

Further improvements in methodology can be made to make the results more accurate. In the current
method, alignment in AutoCAD is done manually, which may cause deviation. Even though iPhone scans
were registered as close as possible to the professional scanned point cloud, the manual operation can still
cause a range of mistakes. Moreover, when measuring surface to surface distance, angled surfaces with
different measure points can result difference in dimensions. Further research should be done to discover
best how to control measuring points. Additionally the fuzzy quality of the smartphone scanned point
clouds increase tracing error caused by human decisions on where to draw the lines.

Even though not as accurate and precise as professional scanner, smartphone scan takes advantage on
being able to handheld. Professional scanners on tripod can have shadow, where laser light cannot reach
therefore creating a blank area in point clouds. By having a smartphone handheld, moving around a target
when scanning, the shadow can be reduced or avoided.

179

Thus, smartphone scanned point clouds are not qualified for professional purposes of heritage
conservation unless further research being conducted to rule out errors caused by human decisions in
measuring procedure. The current research result supports smartphone scanned point clouds as
interpretive drawing, or as assistant document provide alongside hand measurements. The instability
reduced the reliability of smartphone scanning of heritage conservation for professional uses. In case of
endangered heritage and heritage at inaccessible locations, a smartphone scanned documentation can
provide a relatively quick although not as accurate documentation.

5.5 Summary
Chapter 5 exhibited the scanning result of smartphone scans.
5.1 introduced commonly used architectural drawing and digital documentation scale and selected the
standard for the evaluation of iPhone scans.
5.2 mentioned the measurements and result of three methods including smartphone scan, professional
scan, and laser measurements.
5.3 comparisons made; they were divided into two parts, numeric comparison and visual comparison.
Measurement results were compared on each scanning target.
5.4 evaluated the qualification of different modes of smartphone scans.

iPhone acquired scans and professional scanner captured scans were exported from the scanning software
and imported into AutoCAD so that measurements could be made. Potential error from human
interpretation were observed from 1) manual alignment process in measurements, 2) different
measurement setpoints on angled surfaces (walls, ceiling, and floor), 3) distinguishing fuzzy points from
captured surface in point clouds. Sections were created in measure room dimensions. The percentage of
error of each target was calculated. Smartphone scans on furniture scale met architectural drawing

180

standard for heritage conservation purposes. On room or building scale, the quality of data was doubted
with deviation exceeding 1/2 of the smallest unit needed for heritage conservation tasks. Using low point
density mode can achieve a better result with a smaller rate of error, and raw data performing better than
merged files.

Through visual comparison, the result shows smartphone can capture a slightly tilted surface, but the
thickness of points are thicker than professional scanners. iPhone LiDAR scanned architecture detail
include the correct location but lack a clear description of the geometry. The capture and exhibition of
wall intersection and corners from iPhone is a rounded corner instead of a 90 degree angle. Because light
travels in straight lines, professional and smartphone scans taken on tripod will always have shadow in
documentation. Handheld smartphone scan can reduce shadow through moving around targets in
capturing from maximum angle.

With human factors and objective factors impacting scanning result, and improvements to be made, there
is still the possibility for smartphone scanned data qualified for room and building scale architectural
drawings. Suggestions and potential usages of smartphone acquisition on heritage conservation will be
discussed further in Chapter 6.

181

Chapter 6 Qualification
Chapter 6 examines a smartphone scanned data’s application to the four heritage conservation tasks:
documentation, survey, monitoring, and education (Figure 6-1). The accuracy of the data collected by
smartphones are compared against the traditional methods for these tasks. It also looks at the advantages
and disadvantages of using smartphones for heritage conservation tasks and the challenges associated
with it. The chapter also provides recommendations for best practices for using smartphones in
conservation tasks. Finally, the chapter discusses future research directions and technologies that could
further improve the use of smartphones in heritage conservation tasks.

182

Figure 6-1: Chapter 6 overview diagram

183

6.1 Documentation
The qualification of scanned data from smartphones for heritage conservation documentation was based
on the HABS (Historic American Buildings Survey) requirement of paper submissions (Library of
Congress, 2016). The size of the paper and the line weight were considered as impacting factors, which
are dependent on the scales used. As of now, the final output is expected to be a paper drawing with a
specific scale, which can be either generated from point clouds that handed over directly to architects or
directly traced by heritage conservation professionals.

The most common architectural scales for smaller building components, 1/4”=1’-0” and 3/4”=1’-0” were
selected as the standards for the qualification of smartphone scanned data (Table 6-1). Smartphone scans
were manually aligned with professional scanners’ scan in AutoCAD. Data acquired from professional
scanners were considered as ground truth in the comparison (Figure 6-2). Smartphone scans were
measured and compared to the ground truth; errors and error rates were calculated as result. Qualification
standards were determined based on half of the smallest units of the two common architectural scales
(1/2” and 3/16”).

184

Table 6-1: Architectural Scales for drawing (Burns, 2004)
Scale
Smallest
Unit
Evaluating
Unit
Use
1/16" = 1'-0" 4" 2"
Drawings of large structures without
details. Materials shown in plan only.
1/8"=1'-0" 2" 1"
Little detail possible. Materials shown in
plan, only large units in elevation.
1/4"=1'-0" 1" 1/2"
Common architectural scale. Reasonable
amount of detail possible. HABS/HAER
shows door and window frames, materials
in both plan and elevation. At this scale,
line weights can adversely affect
accuracy. A 3x0 (0.25 mm) line is
approximately 1/2" thick.
3/4"=1'-0" 3/8" 3/16"
Common scale for door/window
elevations and other features of similar
scale.
1 1/2"=1'-0" 3/16" 3/32"
Details of door/window jambs/frames,
large tools, small machines, etc.
3"=1'-0" 3/32" 3/64"
Details of objects such as hardware, tools,
etc. and molding profiles.
Full Size
Small or intricate objects, elaborate
moldings and ornamentation

185

Figure 6-2: Smartphone scans were manually aligned with professional scanners’ scan in AutoCAD.
Green point cloud from RTC 360, red point cloud from iPhone

Smartphone scans with SiteScape exhibited a good result in the measurement of furniture, door, and
window size target at 1/4”=1’-0” scale. In scans with either high or low point density mode, errors in
these measurements ranged smaller than 1/2” and bigger than 3/16” (see Chapter 5.3.1). This means
smartphone can be used to measure such targets in a drawing that commonly describe rooms and
floorplans. In the measurement of room dimensions in scale 1/4”=1’-0”, low point density mode exhibited
limited errors in the measurements of two room widths, nevertheless, two room height measurements fall
out of the qualification standards.

186

Horizontal acquisitions were observed as being more accurate than vertical acquisitions in high point
density mode and merged scans, which may because smartphone device was turning horizontally during
acquisition, and the laser can shoot better perpendicularly on vertical walls than roof and floor. When the
laser was shot on a surface with small angle, the accuracy can be reduced, which also happened to
professional scanners. Unfortunately, most of the acquisitions were not accurate enough for scale
3/4”=1’-0” scale, which most commonly used in documenting window and door details.

From visual comparison, smartphone scans can document the thickness of door panels, angled surfaces,
and locations of detail features (see Chapter 5.3.2) (Figures 6-3, 6-4, 6-5). The thickness of bedroom
closet sliding door was accurately described by both professional scanners and smartphone, although with
different expression of points. The slopped roof of bedroom was also recorded by smartphone scans.
Geometries of small features such as lighting fixtures cannot be recorded, but the location can be
accurately described in sections views.

Figure 6-3: Low density iPhone scan (red) and RTC 360 scan (green) closet detail

187

Figure 6-4: High density iPhone scan (red) and RTC 360 scan (green) ceiling detail

Figures 6-5: Low density iPhone scan (red) and RTC 360 scan (green) section 1 (left), and detail (right).

The same as a professional scanners’ limitations, mirrors and glazing are problematic for smartphone to
scan as well. For example, the iPhone laser reflects on a mirror, shoots to the bathroom wall, and is
reflected back, creating a space in the mirror that doesn’t exist in real environment (Figure 6-6). Window
glazing caused laser refraction and deviation on acquisition (Figure 6-7). Thus everything acquired
through mirror and glazing needs to be cropped during data processing.

188

189

Figure 6-6: Mirror reflection captured by smartphone

Figure 6-7: Refracted landscape captured through awning windows in Reunion House living room.

Models can also be traced from scanned point clouds. The accuracy of 3D models generated from point
clouds depends on the quality of the data used. Therefore, it is important to use reliable scanners to
generate the most accurate 3D models. In the case of smartphone scans, the error range of smartphone
scan is laid between 0 inch to 2 5/8" inch within measurements of 2’-5” to 15'-9 1/8" (see Table 5-15 to
Table 5-19). Compared to professional scanners, which can limit error to less than 1/8”, smartphone scans
are not suitable for heritage conservation professionals to use as a reliable source for tracing and creating
3D models that are up to HAB’s standards, but could be used if lower accuracy is acceptable.

Merging scans manually in CloudCompare can create a floor plan of a single family house sized building,
but this process can be prone to errors. While the accuracy of the floor plan is usually satisfactory if the

190

plan is in a loop, its accuracy can suffer if it is linear. Point clouds acquired from professional scanners
also need merging, however, the paid software did a better job than manual alignment and merging.

To serve heritage conservation documentation, smartphone scans can be used in documenting location of
features such as furniture, doors, and windows with architectural scale 1/4”=1’-0”. However, room size
and dimensions are not recommended using smartphone scan. iPhone scans are not accurate enough for
measuring room dimensions in scale 1/4”=1-0”, and for window and door detail drawing with scale
3/4”=1’-0”. With different scanning parameters in SiteScape, low point density achieves better result than
high point density. Raw scan data had better results than merged scans. Scanning angle and path also
impact the accuracy of acquisition. Facing smartphone LiDAR sensor as perpendicular as possible to the
scanning target can help secure accuracy as researched. For creating 3D models as documentation,
smartphone scans are not accurate enough for tracing because the deviations are too big.

6.2 Survey
A smartphone scan is not suitable for generating a site survey for the Reunion House because it has
various limitations that make it difficult to capture and accurately map the outside environment. For the
lot at the Reunion House, too many stiches of scans was required to create a site survey because the
limited range of smartphone scan acquisition. Different from indoor environments that have straight walls
and solid volumes, an exterior environment are more organic and complex. In such environment reference
targets using for scan merging and stitching are not offered, thus impacting quality of survey map.
Furthermore, smartphone LiDAR struggles to capture data accurately in environments with a high tree
canopy, as the LiDAR sensor on a smartphone is not powerful enough to penetrate the gaps between the
leaves, meaning it often misses parts of the environment that need to be surveyed (Figures 6-8, 6-9, 6-10).
This is especially problematic for Reunion House as it is located in a densely wooded area. In addition to

191

the limited range and difficulty in stitching scans, smartphone LiDAR sensors are also susceptible to
interference from external factors. In outdoor environments, the sensor is impacted by glare from the sun,
which can cause distortion in the acquired data and render it unusable.

Figure 6-8: iPhone scanned front yard at Reunion House with a “spider leg” feature.

192

Figure 6-9: “Spider leg” feature detail in iPhone captured point cloud.
Figure 6-10: “Spider leg” feature photograph.

Professional scanners like the RTC 360 are designed with advanced technology that can handle complex
outdoor environments with high tree canopy areas, strong sun light, and other environmental conditions
(Figures 6-11, 6-12). That scanner is equipped with a high-powered laser system that can accurately and
quickly scan the environment with enough amount of strong laser light penetrate tree canopy reaching
ground or building surfaces. The scanner also has multiple sensors that can detect any obstacles in the
environment, such as trees, buildings, and other objects. Additionally, the scanner was more stable under
strong sunlight or insufficient lighting condition.

193

Figure 6-11: RTC 360 works outdoor.

Figure 6-12: RTC 360 scanned outdoor environment at Reunion House

194

Overall, the smartphone scan is not suitable for generating site survey for Reunion House due to its
limited range, difficulty in stitching scans together, susceptibility to interference from external factors,
lack of accuracy, and inability to capture high-resolution data. Professional scanners are equipped with
specialized sensors and can accurately capture large areas with greater accuracy. Aerial surveys can
capture large areas with accuracy and details that are not possible with a smartphone scan. For these
reasons, it is important to use more advanced LiDAR systems when creating site surveys for
environments like Reunion House.

6.3 Monitoring
Building monitoring would track building changes over time. Professional LiDAR scanner can reach a
1/8” accuracy in this task; however, as discussed in 6.1, smartphone scans have larger deviation than
3/16” of inch, which cannot compete with professional scanner’s accuracy. To further validating the
quality of smartphone scanned data, a smartphone scan and a professional scanner scan was aligned in
CloudCompare and computed a cloud to cloud distance, to observe area of deviation (methods were
introduced in 4.3.2.). The result shows most smartphone scan points are in 0 - 0.25 meter (9 inches) from
professional scanner points (Figure 6-13). Red marks deviation larger than 4 inches, which mostly occurs
in blind area of smartphone scans. Green and blue represent deviation from 0 to 4 inches. Points in green
are having larger deviation than points in blue, which more seen on corners and surfaces that laser shoot
not perpendicular on. With most of the surface deviated from the ground truth scan, smartphone scans are
not recommended for heritage conservation monitoring if professionals want to use it for seismic or
detailed documentation because the error from scanning device and movement of building cannot be
distinguished. As mentioned in 6.1 iPhone scans can document door panel thickness, location of small
features, and angle of slopped surface. If these are purpose of monitoring, smartphone scans are
recommended. As the scanning accuracy in smartphones improves, more types of measurements for

195

monitoring will be possible. Although unlikely that very small dimensions will be captured, larger
displacements, for example after a major earthquake, could be feasible.

Figure 6-13: Cloud to cloud distance computation result in CloudCompare

6.4 Education
All scans can be used for digital virtual representation and the circulation of digital cultural reserves. The
iOS application Matterport Capture was used in the case study at Reunion House. The result of scans can
be edited on a smartphone app and uploaded to webpage for viewing. On the webpage, scans can be

196

viewed from doll house view; measurements can be taken online, and virtual tours can be generated for
free (Figure 6-14). A short video walk through and wide-angle photographs are downloadable from the
webpage as well (Figure 6-15). With Matterport, technicians are having less control on the scans that
Matterport processed it and only present the final product.

Figure 6-14: Doll house view of iPhone scanned Reunion House.

197

Figure 6-15: Downloadable Reunion House videos and photos from Matterport webpage.

Other deliverables are available with a cost, such as a schematic floor plan, a point cloud file, or a BIM
file with Matterport scanners (Figure 6-16). According to WCAPT workshop, the quality of point clouds
generated from Matterport scanner was still questionable in the professional field (White, 2023). Just like
Matterport scanner acquired scans, iPhone scans can adjust privacy setting and share publicly. Except for
a lower resolution on camera and unable for other deliverables, iPhone scans perform the same as
Matterport scanners. Thus, considering cost-effective, time, effort, and ability to share and view
universally iPhone scans are high recommended.

198

Figure 6-16: Other deliverables from Matterport webpage.

Although smartphone 3D scans may not meet the professional standards required for survey or
documentation purposes, they can still be useful for education purposes. For example, they can be used to
create 3D models of historical artifacts, buildings, or cultural heritage sites that can be viewed,
manipulated, and studied by students or researchers. A low tolerance for accuracy may be acceptable in
education settings, where the focus is on understanding the general form and structure of the environment,
rather than capturing every detail precisely.

For educational purposes, however, a lower tolerance might be acceptable. Furthermore, smartphone 3D
scanning can also be used as a tool for creative expression and experimentation. For instance, it can be
used by artists or designers to quickly prototype or iterate ideas, or by hobbyists to create 3D models for
interest. Therefore, while smartphone 3D scanning may have limitations, it still has a valuable role to play
in education and other contexts.

199

6.5 Conclusion
Smartphone scans can be used for furniture size object in architecture scale 1/4”=1’-0” drawing for
documentation, monitoring big changes, and for education purposes. iPhone acquisitions were not
accurate enough for professional used in documenting rooms and window and door details in drawings. It
also lacks the ability to capture site information with a high tree canopy and in large scale. Smartphone
scans were also unable to be used in monitoring building movements in seismic monitoring. For these
purposes, smartphone LiDAR cannot be substitute for professional scanners in heritage conservation uses.

While not suitable for certain heritage conservation purposes with current methodology, smartphone
scans can be valuable in many cases. The current product of smartphone scan can be a useful tool for the
documentation of endangered heritage. Heritage 3d scanning in war zones, unreachable sites by larger
equipment, building at risk of immediate changes like those falling down to neglect, or unable to receive
enough attention can be scanned with smartphone by any non-professionals, can still provide valuable
information at lower accuracies than might be ideal. The instability of smartphone scans reduces their
reliability for professional use, but they can be useful for quick documentation in case of endangered
heritage and heritage at inaccessible locations. Further research is needed to address the errors caused by
human decisions in the measuring procedure and to improve the methodology to make smartphone scans
more accurate and reliable.

Smartphone scans can also be used in visual representation of heritage places in presentation to
community discussion, stakeholders, and scholars if provided with supplemental numerical measurement.
It can also be used to document the location of features with additional photograph or scan of details,
which could be beneficial for heritage conservation professionals. The scan can be exported to virtual
reality or augment reality devices, increasing the sense of a place, as well as providing universal access to

200

heritage that have problems with Americans with Disabilities Act. Smartphone scans can also be used to
create insurance record of heritage buildings.

Smartphone scanned point clouds can be manually aligned in AutoCAD. The visual process can increase
deviation then impact measurement result. Furthermore, different from point clouds acquired from a
professional scanner, which have a stable 1/8” thick points layer, smartphone scanned point clouds can be
thicker and fuzzier. The irregularity in thickness and fuzziness can result mistake in human decision on
tracing. It points to the possibility of directly measuring point to point distance from point clouds instead
of tracing in computer aided drawing software. Considering built structures may shift or move over year,
surfaces such as wall, ceiling, and floor can be angled or sloped. A lack of fixed referencing point on non-
regular surface for measurements can cause different reading.

Further research should be done to exempt the possibilities mentioned above and determine acceptable
error rates that might be different from the HABS standard.

201

Chapter 7 Conclusion
Chapter 7 discusses scanning heritage sites with smartphones and future work.

7.1 Scanning heritage sites with smartphones
The present research is focused on developing alternative scanning methods using portable devices for the
purpose of serving conservation goals in the field of heritage conservation. 3D scanning technology plays
a significant role in preserving the physical and geometrical aspects of both tangible and living heritages.
The concept of utilizing smartphones for scanning was inspired by the realization that numerous heritage
sites and buildings are not receiving adequate attention due to disparities in resources. These sites are
often subject to factors such as lack of personnel, insufficient funding, operational barriers, or exclusion
from legal or regulatory lists, which put them at a disadvantage. Therefore, more accessible scanning
tools are required to aid those who are involved in conserving these invaluable cultural assets.

Conventional scanning and data processing methods typically require a high degree of proficiency in
specialized equipment and software, which presents obstacles for individuals in the field of heritage
conservation. The current study proposes to keep the technology accessible to anyone who wishes to scan,
document, and share cultural assets that they consider valuable and significant. It is hoped that this
approach will promote the protection of built environments that are physically inaccessible, endangered,
or ignored.

It is important to emphasize that the use of smartphone scanning is not intended to replace professional
scanned data. Instead, one can identify gaps and limitations in smartphone scanning to gain a better
understanding of its potential applications in the field of heritage conservation and give a comprehensive
analysis of the strengths and limitations of various professional scanners. Scanning has proven to be

202

useful in diverse fields such as medication, industrial manufacturing, and ecosystem monitoring. In the
context of heritage conservation, scans are instrumental in facilitating both digital and physical
reconstruction of objects and sites, monitoring changes over time, evaluating structural states, and
documenting for future research. Together with more traditional techniques such as tape measurements
and photography, scanning provides a critical platform and foundation for researchers and stakeholders to
undertake more effective measures in preserving buildings.

Digital documentation encompasses a range of media and methods, including 2D documentation, 360-
degree photographs, and 3D documentation. Various scanning techniques have been categorized based on
visualization format and data acquiring procedures. These include 2D documentation, encompassing
measured drawings and film, as well as 3D photographs and 3D scanning utilizing various methods
(Figures 7-1, 7-2). 2D scanning techniques commonly include photographs captured with film medium or
panoramic photos, with cylindrical and spherical panoramic photos serving as a subcategorization. In
contrast, 3D spatial data can be acquired through photogrammetry, structured light scan, triangulation
scan, pulse scan, phase-comparison scan, as well as Matterport and 360-degree cameras. Each of these
methods possesses its own strengths and weaknesses, and combining techniques thoughtfully can
maximize benefits and achieve specific goals. With the recent release of iPhones equipped with built-in
LiDAR sensors, the possibility of smartphone scanning assisting heritage conservation professionals in
their work has emerged. However, limited attention has been paid to smartphone scanning in academia.
Several journals have examined the accuracy of smartphone scanning, but this needs to be evaluated on a
case-by-case basis. Further research is needed in this area, and to that end, permission has been obtained
from the Neutra Institute to use the Reunion House at Silver Lake as a case study.

203

Figure 7-1: 360 degree photograph of Reunion House master bedroom.

Figure 7-2: 3D scan of Reunion House master bedroom.

204

Richard Neutra is widely regarded as one of the most significant architects of the 20th century, renowned
for his international style practice in the United States. The Reunion House at Silverlake, Los Angeles,
represents a culmination of Neutra's signature architectural elements and was designed programmatically
as a house featuring separate quarters for grandparents and grandchildren, along with a central meeting
space (Lamprecht, 2021). This property is currently owned by the Neutra Institute for Survival Through
Design, and was designated as a Historic-Cultural Monument by the City of Los Angeles in 2021.

Figure 7-3: Historic photograph of Reunion House (University of Southern California, 1951)

A methodology for scanning heritage sites using a smartphone has been developed (Fig. 7-4). The study
employed an iPhone 13 Pro and two smartphone apps, SiteScape and Matterport Capture, for data
acquisition and processing. Four key heritage conservation tasks, namely site survey, documentation,
monitoring, and education, were identified for validating the capability of smartphone scanning. The

205

proposed methodology included the selection of appropriate software, data acquisition, data processing,
comparison to traditional methods, and a case study involving professional scanners. By following this
methodology, one can utilize a smartphone and select suitable software to scan a heritage site in order to
achieve heritage conservation goals. SiteScape and Matterport smartphone apps were chosen due to their
exceptional ability to fulfill the research objectives. SiteScape was designed for use in data acquisition
processes such as surveying, documenting, and monitoring. The desktop software CloudCompare was be
utilized for registering multiple scans, thereby enhancing accuracy and precision. Furthermore,
smartphone-scanned data was compared to traditional methods and professional scanner-scanned data.
Matterport was selected for the purpose of its potential for educational uses. The data acquisition,
processing, and visualization achieved through Matterport were compared to traditional methods and a
high-end scanner.

206

Figure 3-1 Proposed methodology

207

Test scans were conducted at the Hoose Library of Philosophy in Mudd Hall, USC campus (Figure 7-4).
A detailed start up guide was written as instructions for heritage conservation professionals. The start-up
guide describes the process of scanning the interior of a space using SiteScape and merging multiple
scans to generate a floor plan. It also outlines the process of scanning the same area with different point
density parameters (low, medium, and high) using SiteScape. Two of the three scans were used to
investigate the possibility of overlaying scans to increase accuracy. It introduced the process of acquiring
geometric data of the same area using SiteScape with the same parameter. CloudCompare was utilized to
compute the distance between the two exact scans (scan 3 and scan 4) to examine the repeatability of
smartphone-scanned data. The last part of the startup guide demonstrated Matterport Capture for iPhone
and a tripod to take 360-degree photographs to generate a virtual tour that can be shared through links on
a website.

Figure 7-4: Hoose Library of Philosophy at Mudd Hall, USC.

208

Figure 7-5: Point cloud scan of Hoose Library of Philosophy.

The results from the test scans demonstrated the relative simplicity of all four methodologies. The floor
plan generated in the first methodology could be used for site survey purposes and a comparison of this
floor plan with scanner-scanned data may offer valuable insights (Figure 7-6). The second methodology
shows that overlapping two scans can increase the points describing the same surface, but a thicker layer
of points may not contribute to documentation due to increased error (Table 7-1). For monitoring
purposes, the third methodology shows that smartphone-scanned data cannot produce repetitive data
every time unless more stable scanning methods or better alignment computation is employed (Figure 7-
7).

209

Figure 7-6: Merged scan of Hoose Library of Philosophy at Mudd Hall through CloudCompare.

Table 7-1: Result in number of total points from point cloud from overlapping.
Scans Number of Points Point Cleared
Scan 1 Filter 2,764,678
785,812 Scan 1 3,550,490
Scan 2 Filter 5,540,899
227,139 Scan 2 5,768,038
Scan 1+2 Merge 9,318,528

Scan 1+2 Filter-Crop-Merge 8,305,577 1,012,951
Scan 1+2 Merge-Filter-Crop 7,091,553 2,226,975

210

Figure 7-7: Cloud to cloud distance computation result in CloudCompare

The study conducted at Reunion House involved a comparison of smartphone scans and professional
scans in the master bedroom. Both sets of scans were imported into AutoCAD, and their respective
dimensions were compared, following which a range of errors was computed. The comparison results
were subsequently compiled in a tabular format, and recommendations were provided based on the
outcomes of the analysis. The scale and standard of accuracy are important when generating drawings and
models for heritage conservation. Architectural scales, such as 1/4”=1’-0” and 3/4”=1’-0”, are commonly
used to document features of similar scale, with the smallest unit being 1” and 3/8”, respectively. To
evaluate the performance of smartphone scans in creating as-built drawings on paper, scales of 1/4”=1’-0”
and 3/4”=1’-0” with the smallest units of 1/2” or 3/16” will be used as a standard. For digital
documentation, a standard accuracy of 1/8” is required for tracing three-dimensional models, floor plans,
or section drawings. The most common scale used for engineering drawings and maps is 1”=20’, with the
smallest unit being 0.4’. To determine 0.4’ in a map, a factor of two than the smallest feature, which is
0.2’ (2 2/5”), is needed.

211

A final comparison was to tests the accuracy of smartphone scanned data acquired on a tripod to
professional scanner scanned data (Figure 7-5). The smartphone scans are limited by the angle of the
LiDAR sensor, resulting in blind areas on the ceiling and floor. Tilting the smartphone up or down can
reduce the blind areas. The scans were manually aligned with professional scanner captured point clouds,
and measurements were taken using tracing and the linear dimension tool. The thickness and fuzziness of
points in the iPhone scans varied based on different point density, distance, and captured angle, making
tracing and measurement difficult. Room dimensions were measured individually in all eight scans
through the plans and sections, but the width of one room could not be measured due to missing data.

Figure 7-5: Overlapping professional scanner point cloud with iPhone scanned point cloud.

212

Three section views were created to compare the deviations and differences between the two devices.
Through visual comparison, smartphone scans can capture details such as the thickness of sliding doors,
slop of roof, and location of features.

Potential errors arising from human interpretation can occur during the manual alignment process,
different measurement setpoints on angled surfaces such as walls, ceiling, and floor, and distinguishing
fuzzy points from the captured surface in point clouds. In order to assess the accuracy of the scans,
sections were created to measure room dimensions, and the percentage of error of each target was
calculated.

Smartphone scans on furniture scale met the architectural drawing standard and can be used for heritage
conservation purposes. However, for room or building scale, the quality of data was questionable with
deviation exceeding 1/2 of the smallest unit needed for heritage conservation tasks. The use of low point
density mode was found to achieve a better result with a smaller rate of error, with raw data performing
better than merged files. Overall, the findings suggest that while smartphone scans may be suitable for
heritage conservation tasks, there are limitations to their accuracy and reliability, and further
improvements in methodology are necessary to improve their effectiveness for professional use.

Qualifications of heritage conservation tasks including site survey, documentation, monitoring, and
education were made (Figure 7-). Smartphone scans are not a suitable option for site surveys due to
limitations imposed by the vegetation canopy and lighting conditions, which necessitate excessive
merging of scans. However, the high point density mode of smartphone scanning is deemed appropriate
for horizontal acquisition in documentation, particularly at a scale of 3/4"=1'-0". In contrast, it is not
recommended for the documentation of room sizes, as it is unable to capture details accurately.
Furthermore, smartphone scanning is not recommended for monitoring building shifts, as the deviation is
too large to distinguish between scanning errors and actual building movement. Despite these limitations,

213

smartphone scanning can be useful for educational purposes, as it can yield results similar to those
obtained from professional scanners such as Matterport. Furthermore, the strengths of smartphone
scanning include its cost-effectiveness, ease of use, and universal shareability. However, these
applications must be approached with caution, as the limitations of smartphone scanning can lead to
errors and inaccuracies.

Figure 7-6: Qualification of smartphone scan for heritage conservation tasks.

7.2 Future work
Although the goal of proposing a methodology of conducting scans with smartphones for heritage
conservation was accomplished, there are several improvements that could be done in order to achieve
better results.
7.2.1 Improvements

214

Many adaptations, tests, and experiments that could have been beneficial were left for future studies due
to a lack of time. These include increasing scan features and trials in the case study to produce an
additional list of products from smartphone scanning, which can potentially lead to a better representation
of heritage resources. Larger-scale heritage sits can also be merged with supplemental point clouds
acquired from smartphones. Moreover, the potential of the point cloud data can be further investigated for
other heritage conservation purposes. Smartphone scanned data can be computed in different software
using greater variety of tools in evaluating quality and qualifications. Scans obtained from different
devices, techniques, and software such as Android, photogrammetry, and Polycam can be compared with
the existing data to provide a more comprehensive understanding of the capability of the smartphone as a
portable and affordable scanning device for heritage conservation professionals. With more time,
improvements can be done by implementing smartphone-captured data into game engines to simulate
experiences in virtual environments with augmented reality and virtual reality (Figures 7-7, 7-8).
Furthermore, the scanned point cloud can be computed into a mesh and 3D printed on different scales,
exhibited at different locations, or mass-produced. This would help to increase the level of understanding
of heritage sites and also could create a sense of connection with the history.

215

Figure 7-7: Game engine used in architecture (The Architect’s Newspaper, 2019)

Figure 7-8: Virtual reality in architecture (Parametric Architecture, 2022)

216

For the purpose of benefiting heritage conservation professionals with accessible tools, improvements in
the clarity and details of the getting start guide (Chapter 4) can be helpful. The guide can also be filmed
as a video to engage with diverse audiences. Publishing the guide together with the research outcome in
journals and online will make smartphone scans available for broader heritage conservation professionals.
Presenting at conferences and workshops will have a similar positive effect. The use of smartphones as a
tool to obtain data offers the potential to expand the accessibility of heritage to the public, making
heritage more accessible, interesting, and engaging. Online platforms such as Sketchfab allow users to
publish and share 3D content; if the methodology can be widely adopted, the point clouds should be
uploaded and published.

7.2.2 Other topic areas
There are other areas in heritage conservation that could benefit from the use of the smartphone. Heritage
conservation is a broad field that involves various tasks beyond the areas considered: site survey,
documentation, monitoring, and education (Figure 7-9). Although the proposed tasks are crucial, heritage
conservation encompasses many other purposes that are equally essential. Smartphone capabilities in the
field can be explored more extensively to aid in the conservation of heritage.

217

Figure 7-9: Heritage conservation task diagram.

Not all smartphones have a LiDAR sensor, which is an essential tool for creating 3D models of objects or
structures. However, there are many other ways a smartphone can assist heritage conservation
professionals in reducing costs. For example, the camera function of a smartphone can be used to capture
images of objects or structures that need to be conserved or restored in creating records. These images can
then be used for reference purposes during the conservation process. Moreover, the recording function on
a smartphone can be used to capture audio recordings of interviews with experts or oral histories related
to heritage objects or structures in documenting oral history. This information can then be used to
understand the cultural significance of the object or structure and inform the conservation process.
Smartphone applications such as Fulcrum help create an architecture description in an organized way.
With smartphones, text or audio information can be scanned through a QR code and delivered to an

218

audience in exhibitions. Translate audio to text, document scan, and GPS system on smartphone can all
benefit cultural heritage protection and documentation.

Additionally, there are various apps and software available on smartphones that can aid heritage
conservation professionals in their work. For example, there are apps that can be used to measure
distances and angles accurately, which can be helpful in creating precise 3D models of objects or
structures. The built-in functions of a smartphone can be beneficial for preliminary jobs in heritage
conservation tasks, and more research can be done in this area. Although not all smartphones have the
same capabilities, there are still many ways that a smartphone can assist heritage conservation
professionals in reducing costs and making their work more efficient. The current capabilities of
smartphone LiDAR are still limited, and the accuracy and resolution of scans can vary widely. If
smartphone LiDAR technology were to improve, it could greatly enhance the possibility of using
smartphones for heritage conservation, making it easier to document and preserve historic sites and
artifacts.

Additionally, more research on photogrammetry, which is another technique used to create 3D models,
could also lead to further advancements in smartphone scanning for heritage conservation. By combining
these technologies and improving their capabilities, the potential for preserving cultural heritage could be
greatly increased. Therefore, exploring the capabilities of smartphones in the field of heritage
conservation can lead to new and innovative ways to preserve our cultural heritage.

7.2.3 Future work
During the research, obstacles appeared in the alignment and distance computing of smartphone-scanned
point clouds. Without improvements in the algorithm of alignment and true distance computing from
CloudCompare, the accuracy of data cannot be validated. Future work can be done to either improve the

219

algorithm or use another software to process the data. More research and tests should be done to
structures and sites with various scales. With the collective data, benchmarks and metrics can be produced
for scanning with smartphones. A multiplicity of supported equipment such as drones should also be
implemented in future works for sites or structures that cannot be easily accessed. The limits and strength
of smartphones as 3D scanning devices for heritage conservation should also be tested in endangered or
hostile environments.

With cost-effectiveness being a concern for certain projects, alternative low-cost scanning solutions such
as photogrammetry and structure from motion should be considered. Results from different cost-effective
documentation methodologies should be compared. The potential of smartphones in heritage conservation
should be expanded too. The acoustics of structures has been studied by scholars; based on the research,
smartphone application in identifying acoustic differences among heritage sites can be developed. The
public can learn the structure from a different perspective. The application can also create a more
equitable environment for people with vision disabilities. 3d scanning with 3d printing could also be
done.

7.3 Summary
The use of a smartphone allows for accessibility and affordability of 3D scanning to heritage conservation
professions over that provided by more expensive professional solutions. However, it is not the only issue
to face. HABS determined digital technologies are more suitable as tools for generating documentation of
structures than the final product because the frequently updated software and hardware cause data
migration and archival issues. The future of point clouds acquired in the past decades should be a
concern. The possibility of an official file format that can be shared universally and stored locally with a
small file size for decades should be considered in future research. With different researchers and
organizations producing results of varying quality, the 3D scanning process is often inconsistent.

220

Developing standards for the amount and quality of data to be collected is necessary to ensure that the
data being collected is consistently high quality and can be used for accurate comparison and analysis.

Point clouds can be traced in computer-aided modeling and drawing software, which eventually required
decision-making to translate the point clouds into a product. If in the future 3D scanned data can be the
format for the architecture, engineering, and construction industry, the deviation can be further reduced,
and accuracy can increase. In the unforeseen future, point clouds may replace modeling from software
such as Revit and Rhino; materials and properties can be directly assigned to the cluster of points. There
are current software and websites generating 2D drawings directly from point clouds, nevertheless, the
algorithm cannot process certain complicated geometry. The advancement can also be a direction for
future research.

Another area of future research is the use of smartphone scanning to evaluate the emotional impact of
heritage sites. By creating 3D models of heritage sites, researchers can analyze the physical features that
contribute to a sense of place and feelings. This information can be used to identify ways to enhance the
emotional impact of heritage sites and create more meaningful experiences for visitors. Social media
applications might be able to combine several of the afore-mentioned techniques As the technology on
smartphones improves, heritage conservationists have more opportunities towards augmenting their
existing toolset.

221

Bibliography
A Friends of the Michael White Adobe. (2022). The Michael White adobe. Retrieved April 26 from
http://www.michaelwhiteadobe.com/

Acuity Laser. (2022). Principles of measurement used by laser sensors and scanners. Acuity Laser. Retrieved April
26 from https://www.acuitylaser.com/sensor-resources/measurement-principles/

Agnello, F., Lo Brutto, M., & Lo Meo, G. (2005, 26 September – 01 October). DSM and digital orthophotos in
cultural heritage documentation. CIPA 2005 XX International Symposium, Torino, Italy.

Akboy-İlk, S. (2017). Crafting the architectural measured drawings. The Plan Journal, 2(1), 39-61.
https://doi.org/10.15274/tpj.2017.02.01.04

Alshawabkeh, Y., Baik, A., & Fallatah, A. (2021). As-textured as-built BIM using sensor fusion, zee Ain historical
village as a case study. Remote Sensing, 13(24), 5135. https://doi.org/10.3390/rs13245135

Andrews, D. P., Beckett, N. J., Clowes, M., & Tovey, S. M. (2003). Photographic survey of mosaic and tiled floors–
A methodology. International Archives of the Photogrammetry, Remote Sensing and Spatial Information
Sciences, 34(Part XXX), 1-6.
https://www.academia.edu/download/32243647/Photographic_Survey_of_Mosaic_and_Tiled_Floors_-
_a_methodlolgy.pdf

Arbogast, D. (2010). How to write a historic structure report. WW Norton & Company.

Argyriou, L., Economou, D., & Bouki, V. (2020). Design methodology for 360 immersive video applications: the
case study of a cultural heritage virtual tour. Personal and Ubiquitous Computing, 24, 843-859.
https://doi.org/10.1007/s00779-020-01373-8

Arias, P., Herraez, J., Lorenzo, H., & Ordóñez, C. (2005). Control of structural problems in cultural heritage
monuments using close-range photogrammetry and computer methods. Computers & Structures, 83(21-
22), 1754-1766. https://doi.org/10.1016/j.compstruc.2005.02.018

Attarian, K., & Safar Ali Najar, B. (2022). Heritage documentation and structural analysis of historic water-supply
canals. Environmental Earth Sciences, 81(3), 100. https://doi.org/10.1007/s12665-022-10206-1

Autodesk. (2022). What’s new in ReCap Pro. Autodesk. Retrieved April 26 from
https://www.autodesk.com/products/recap/features

Bahadursingh, N. (2021, June 11). Neutra reunion house added to Los Angeles’ list of historic-cultural monuments.
Archinect. Retrieved April 26 from https://archinect.com/news/article/150267554/neutra-reunion-house-
added-to-los-angeles-list-of-historic-cultural-monuments

Bak, J. (2017, 13 March). 3D photograph. Retrieved Apirl 26 from https://unsplash.com/photos/mOkQGzaRvco

Ballarin, M., Balletti, C., & Vernier, P. (2018, 4-7 June). Replicas in cultural heritage: 3D printing and the museum
experience. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information
Sciences, Riva del Garda, Italy.

Bent, G. R., Pfaff, D., Brooks, M., Radpour, R., & Delaney, J. (2022). A practical workflow for the 3D
reconstruction of complex historic sites and their decorative interiors: Florence As It Was and the church of
Orsanmichele. Heritage Science, 10(1), 1-17. https://doi.org/10.1186/s40494-022-00750-1

222

Bharadwaj, S., Dubey, R., Zafar, M. I., Tiwary, S. K., Faridi, R. A., & Biswas, S. (2022). A novel method to
determine the optimal location for a cellular tower by using LiDAR data. Applied System Innovation, 5(2),
30. https://doi.org/10.3390/asi5020030

Bitfab. (2022). 3D scanning with structured light. Bitfab. Retrieved Aapril 26 from https://bitfab.io/blog/3d-
structured-light-scanning/

Block, F. (1951). Reunion house, Silver Lake, Los Angeles, Calif., 1951 [Image]. [Interior photograph of the living
room, view onto the lake, Reunion house (later the residence of Dion and Lynn Smart Neutra), 2240 Earl
Street (aka 2240 Neutra Place), Silver Lake, Los Angeles, California, 1951. Designed by Richard Neutra.].
University of Southern California. Libraries, Los Angeles, USA.
https://calisphere.org/item/cbee092da9cccb48f4ed59ce42bfe985/

Boardman, C., & Bryan, P. (2018). 3D laser scanning for heritage: Advice and guidance on the use of laser
scanning in archaeology and architecture. Historic England.

Burns, J. A. (2003). Recording historic structures (2nd ed.). John Wiley & Sons.

Caine, M. (2019). Conegliano Italian Synagogue - Point Cloud. Sketchfab. Retrieved April 26 from
https://sketchfab.com/3d-models/conegliano-italian-synagogue-point-cloud-
e8cadf05fe7e40fdafb5da1b1b1b4937

CloudCompare. (2022). CloudCompare 3D point cloud and mesh processing software open source project.
CloudCompare. Retrieved April 26 from http://www.cloudcompare.org

Crabbe, L. (2013, December 09). App turns your smartphone into a 3D scanne. DPReview. Retrieved April 26 from
https://www.dpreview.com/news/7548339892/app-smartphone-3d-scanner

d'Annibale, E., Tassetti, A. N., & Malinverni, E. S. (2013). From panoramic photos to a low-cost photogrammetric
workflow for cultural heritage 3D documentation. The International Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences, 40, 213-218. https://doi.org/10.5194/isprsarchives-XL-
5-W2-213-2013

Daly, P., Feener, R. M., Ishikawa, N., Mujah, I., Irawani, M., Hegyi, A., Baranyai, K., Majewski, J., & Horton, B.
(2022). Challenges of Managing Maritime Cultural Heritage in Asia in the Face of Climate Change.
Climate, 10(6), 79. https://doi.org/10.3390/cli10060079

Daneshmand, M., Helmi, A., Avots, E., Noroozi, F., Alisinanoglu, F., Arslan, H. S., Gorbova, J., Haamer, R. E.,
Ozcinar, C., & Anbarjafari, G. (2018). 3d scanning: A comprehensive survey. arXiv preprint
arXiv:1801.08863, 1-18. https://doi.org/10.48550/arXiv.1801.08863

Douglass, M., Lin, S., & Chodoronek, M. (2015). The application of 3D photogrammetry for in-field documentation
of archaeological features. Advances in Archaeological Practice, 3(2), 136-152.
https://doi.org/10.7183/2326-3768.3.2.136

Douglass, M. J. J. (2022). Can optical scanning technologies replace CT for 3D printed medical devices in radiation
oncology? Journal of Medical Radiation Sciences, 69(2), 139-142. https://doi.org/10.1002/jmrs.579

Duluth Preservation Alliance. (2022). What is preservation? Duluth Preservation Alliance. Retrieved April 26 from
https://duluthpreservation.org/what-is-preservation/

El-Ashmawy, N., & Shaker, A. (2014). Raster vs. point cloud LiDAR data classification. International Archives of
the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, 40(7), 79-83.
https://doi.org/10.5194/isprsarchives-XL-7-79-2014

223

Eppich, R., & Chabbi, A. (2007). Recording, documentation and information management for the conservation of
heritage places: illustrated examples. The Getty Conservation Institute.

Esaias, O., Noonan, G. W., Everist, S., Roberts, M., Thompson, C., & Krosch, M. N. (2020). Improved area of
origin estimation for bloodstain pattern analysis using 3D scanning. Journal of Forensic Sciences, 65(3),
722-728. https://doi.org/10.1111/1556-4029.14250

Existing Condition. (2022). Understanding the hardware: LIDAR & phase-based scanning. Existing Condition.
Retrieved April 26 from https://www.existingconditions.com/knowledge/understanding-the-hardware-lidar-
phase-based-scanning

Exploring Our Fluid Earth. (2022). Practices of science: Precision vs. accuracy. Exploring Our Fluid Earth.
Retrieved April 26 from https://manoa.hawaii.edu/exploringourfluidearth/physical/world-ocean/map-
distortion/practices-science-precision-vs-accuracy

Fais, S., Cuccuru, F., Casula, G., Bianchi, M. G., & Ligas, P. (2019). Characterization of rock samples by a high-
resolution multi-technique non-invasive approach. Minerals, 9(11), 664.
https://doi.org/10.3390/min9110664

Fais, S., Cuccuru, F., Ligas, P., Casula, G., & Bianchi, M. G. (2017). Integrated ultrasonic, laser scanning and
petrographical characterisation of carbonate building materials on an architectural structure of a historic
building. Bulletin of Engineering Geology and the Environment, 76(1), 71-84.
https://doi.org/10.1007/s10064-015-0815-9

Faro. (2022a). FARO focus laser scanners: Capture accurate and complete measurements of buildings, facilities
and complex environments with confidence. Faro. Retrieved April 26 from
https://www.faro.com/en/Products/Hardware/Focus-Laser-Scanners

Faro. (2022b). Faro software solutions. Faro. Retrieved April 26 from https://www.faro.com/en/Products/Software

Faro. (2022c). Understanding laser scanners. Faro. Retrieved April 26 from https://www.faro.com/en/Resource-
Library/Article/understanding-laser-scanners

Filgueira, A., González-Jorge, H., Lagüela, S., Díaz-Vilariño, L., & Arias, P. (2017). Quantifying the influence of
rain in LiDAR performance. Measurement, 95, 143-148.
https://doi.org/10.1016/j.measurement.2016.10.009

Filisetti, A., Marouchos, A., Martini, A., Martin, T., & Collings, S. (2018, 22-25 October). Developments and
applications of underwater LiDAR systems in support of marine science. OCEANS 2018 MTS/IEEE
Charleston, Charleston, SC, USA.

Garashchenko, Y., Kogan, I., & Rucki, M. (2022). Comparative accuracy analysis of triangulated surface models of
a fossil skull digitized with various optic devices. Metrology and Measurement Systems, 29(1), 37-51.
https://doi.org/10.24425/mms.2022.138547

Gargoum, S., Karsten, L., El-Basyouny, K., & Chen, X. (2022). Enriching roadside safety assessments using LiDAR
technology: disaggregate collision-level data fusion and analysis. Infrastructures, 7(1), 7.
https://doi.org/10.3390/infrastructures7010007

Geo Week News Staff. (2004, June 22). Time-of-flight vs. phase-based laser scanners: Right tool for the job. Geo
Week News. https://www.geoweeknews.com/news/time-of-flight-vs-phase-based-laser-scanners-right-tool-
for-the-job

Giraudot, S. (2022). Surface reconstruction from point clouds. Computational Geometry Algorithms Library.

224

Gitbook. (2022). Share 3D guidelines: Active systems: long to mid range. Gitbook. Retrieved April 26 from
https://carare.gitbook.io/share-3d-guidelines/3d-process/data-capture/long-range-techniques

Gonçalves, G., Santos, S., Duarte, D., & Gomes, J. (2019, May 03-05 ). Monitoring local shoreline changes by
integrating UASS, airborne LiDAR, historical images and orthophotos. GISTAM 2019 - Proceedings of the
5th International Conference on Geographical Information Systems Theory, Applications and Management,
Heraklion, Crete, Greece.

Gray, C. (2022). Personal communication.

Grifoni, E., Gargano, M., Melada, J., Interlenghi, M., Castiglioni, I., di Sturmeck, S. R. G., & Ludwig, N. (2022,
October 20-22). Documenting Cultural Heritage in very hostile fruition contexts: the synoptic visualization
of Giottesque frescoes by Multispectral and 3D Close-range Imaging.Journal of Physics: Conference Series
2021 International Conference on Metrology for Archaeology and Cultural Heritage (MetroArchaeo 2021),
Milan, Italy.

Haddad, N. A. (2011). From ground surveying to 3D laser scanner: A review of techniques used for spatial
documentation of historic sites. Journal of King Saud University-Engineering Sciences, 23(2), 109-118.
https://doi.org/10.1016/j.jksues.2011.03.001

Harbowo, D., Sitorus, J., Agustina, L., Muztaba, R., Julian, T., & Malasan, H. (2022, December 08-09). 3D
modelling of meteorite from Astomulyo Village, Lampung, Indonesia by close range photogrammetry
(CRP) methods.IOP Conference Series: Earth and Environmental Science 1st International Seminar on
Earth Sciences and Technology, Online.

Hirtle, N. O., Stepanuk, J. E. F., Heywood, E. I., Christiansen, F., & Thorne, L. H. (2022). Integrating 3D models
with morphometric measurements to improve volumetric estimates in marine mammals. Methods in
Ecology and Evolution, 13(11), 2478-2490. https://doi.org/10.1111/2041-210X.13962

Holcomb, A. (2021). Mobile Phone depth sensors for forest carbon measurement [Master's thesis, University of
Waterloo]. Waterloo. http://hdl.handle.net/10012/17043http://hdl.handle.net/10012/17043

Ikkala, L., Ronkanen, A. K., Ilmonen, J., Similä, M., Rehell, S., Kumpula, T., Päkkilä, L., Klöve, B., & Marttila, H.
(2022). Unmanned aircraft system (UAS) structure-from-motion (SfM) for monitoring the changed flow
paths and wetness in minerotrophic peatland restoration. Remote Sensing, 14(13), 3169.
https://doi.org/10.3390/rs14133169

Jaboyedoff, M., Oppikofer, T., Abellán, A., Derron, M.-H., Loye, A., Metzger, R., & Pedrazzini, A. (2012). Use of
LIDAR in landslide investigations: A review. Natural Hazards, 61, 5-28. https://doi.org/10.1007/s11069-
010-9634-2

Jokilehto, J. (2021). Question of integrity. Change Over Times, 10(2), 94-183.

Jovančević, I., Pham, H. H., Orteu, J. J., Gilblas, R., Harvent, J., Maurice, X., & Brèthes, L. (2017). 3D point cloud
analysis for detection and characterization of defects on airplane exterior surface. Journal of
Nondestructive Evaluation, 36(4), 74. https://doi.org/10.1007/s10921-017-0453-1

Karganroudi, S. S., Aminzadeh, A., Ibrahim, H., Rahmatabadi, D., Francois, V., & Cuillière, J.-C. (2023). A novel
automated approach for geometric reconstruction and flexible remanufacturing of spur gears using point
cloud mapping analysis. Computer-Aided Design and Applications, 92-108.
https://doi.org/10.14733/cadaps.2023.92-108

Koehl, M., Schneider, A., Fritsch, E., Fritsch, F., Rachedi, A., & Guillemin, S. (2013). Documentation of historical
building via virtual tour: the complex building of baths in Strasbourg. The International Archives of the
Photogrammetry, Remote Sensing and Spatial Information Sciences, 40, 385-390.
https://doi.org/10.5194/isprsarchives-XL-5-W2-385-2013

225

Lamprecht, B. (2020). Dion Neutra/ Reunion house historic-Cultural monument application.

Lamprecht, B. (2021). Dion Neutra / Reunion house historic context, architectural description, statement of
significance, and bibliography.

Lauria, G., Sineo, L., & Ficarra, S. (2022). A detailed method for creating digital 3D models of human crania: an
example of close-range photogrammetry based on the use of structure-from-motion (SfM) in virtual
anthropology. Archaeological and Anthropological Sciences, 14(3), 42. https://doi.org/10.1007/s12520-
022-01502-9

Leica Geosystems AG. (2022a). Leica cyclone 3D point cloud processing software. Leica Geosystems AG.
Retrieved April 23 from https://leica-geosystems.com/products/laser-scanners/software/leica-cyclone

Leica Geosystems AG. (2022b). Leica ScanStation P30/P40. Leica Geosystems AG. Retrieved April 26 from
https://ps.leica.gs/wp-content/uploads/2019/07/Leica_ScanStation_P30-P40_Plant_DS.pdf

Leica Geosystems AG. (2022c). Leica ScanStation P30/P40 product specifications. Leica Geosystems AG.
Retrieved April 23 from

Library of Congress. (2011). Heritage documentation programs HABS/HAER/HALS photography guidelines
[Documentation programs]. L. o. Congress. http://www.loc.gov/pictures/collection/hh/

Library of Congress. (2016, April 30). Historic American Buildings Survey (HABS). Library of Congress. Retrieved
October 19 from https://www.nps.gov/hdp/habs/

Lievendag, N. (2017, March 15). HP 3d structured light scanner pro s3 (David SLS-3) review. 3DScanExpert.
Retrieved April 26 from https://3dscanexpert.com/hp-3d-scanner-pro-s3-david-sls-3-review/

Liu, Z., Brigham, R., Long, E. R., Wilson, L., Frost, A., Orr, S. A., & Grau-Bové, J. (2022). Semantic segmentation
and photogrammetry of crowdsourced images to monitor historic facades. Heritage Science, 10(1), 1-17.
https://doi.org/10.1186/s40494-022-00664-y

Los Angeles Department of City Planning. (2008). Los Angeles Department of City Planning: Recommendation
report. https://www.docin.com/p-1935806496.html

Luetzenburg, G., Kroon, A., & Bjørk, A. A. (2021). Evaluation of the Apple iPhone 12 Pro LiDAR for an
application in geosciences. Scientific Reports, 11(1), 22221. https://doi.org/10.1038/s41598-021-01763-9

Masciotta, M. G., Sanchez-Aparicio, L. J., Oliveira, D. V., & Gonzalez-Aguilera, D. (2023). Integration of laser
scanning technologies and 360º photography for the digital documentation and management of cultural
heritage buildings. International Journal of Architectural Heritage, 17(1), 56-75.
https://doi.org/10.1080/15583058.2022.2069062

McPherron, S. P., Gernat, T., & Hublin, J.-J. (2009). Structured light scanning for high-resolution documentation of
in situ archaeological finds. Journal of Archaeological Science, 36(1), 19-24.
https://doi.org/10.1016/j.jas.2008.06.028

MeshLab. (2022). MeshLab. Retrieved April 26 from https://www.meshlab.net

Mill, T. (2020). Estimation of Accuracy and Reliability of Terrestrial Laser Scanner in the Detection of Object
Shape. Baltic Journal of Modern Computing, 8(2), 337-346. https://doi.org/10.22364/BJMC.2020.8.2.09

Mineo, S., Caliò, D., & Pappalardo, G. (2022). UAV-based photogrammetry and infrared thermography applied to
rock mass survey for geomechanical purposes. Remote Sensing, 14(3), 473.
https://doi.org/10.3390/rs14030473

226

Montusiewicz, J., Miłosz, M., Kęsik, J., & Żyła, K. (2021). Structured-light 3D scanning of exhibited historical
clothing—A first-ever methodical trial and its results. Heritage Science, 9(1), 74.
https://doi.org/10.1186/s40494-021-00544-x

Mount Vernon Ladies’ Association. (2022, October 30). Historic Preservation. Mount Vernon. Retrieved April 26
from https://www.mountvernon.org/preservation/architecture/

Murtiyoso, A., Grussenmeyer, P., Landes, T., & Macher, H. (2021). First assessments into the use of commercial-
grade solid state lidar for low cost heritage documentation. International Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences - ISPRS Archives, 43(B2-2021), 99-604.
https://doi.org/10.5194/isprs-archives-XLIII-B2-2021-599-2021

Naranjo, J. M., Parrilla, Á., & de Sanjosé, J. J. (2018). Geometric characterization and interactive 3D visualization
of historical and cultural heritage in the province of Cáceres (Spain). Virtual Archaeology Review, 9(18), 1-
11. https://doi.org/10.4995/var.2018.6010

Nasiri, V., Darvishsefat, A. A., Arefi, H., Griess, V. C., Sadeghi, S. M. M., & Borz, S. A. (2022). Modeling forest
canopy cover: A synergistic use of Sentinel-2, aerial photogrammetry data, and machine learning. Remote
Sensing, 14(6), 1453. https://doi.org/10.3390/rs14061453

National Park Service. The treatment of historic properties. National Park Service. Retrieved October 29 from
https://www.nps.gov/subjects/historicpreservation/standards.htm

National Park Service. (2021, May). What is historic preservation? National Park Service. Retrieved April 26 from
https://www.nps.gov/subjects/historicpreservation/what-is-historic-preservation.htm

National Park Service. (2022a). Historic structure report format.
https://ohp.parks.ca.gov/pages/1069/files/historic%20structure%20report%20format.pdf

National Park Service. (2022b). The secretary of the interior’s standards for the treatment of historic properties.
https://www.nps.gov/orgs/1739/secretary-standards-treatment-historic-properties.htm

Norman, T., Ilene, R. T., & Ted, J. L. (2018a). Designation of historic places. In Historic preservation: an
introduction to its history, principles, and practice (3rd ed., pp. 95-103). W.W. Norton & Company.

Norman, T., Ilene, R. T., & Ted, J. L. (2018b). Preservation planning. In Historic preservation: an introduction to
its history, principles, and practice (3rd ed., pp. 275-277). W.W. Norton & Company.

Nowak, R., Orłowicz, R., & Rutkowski, R. (2020). Use of TLS (LiDAR) for building diagnostics with the example
of a historic building in Karlino. Buildings, 10(2), 24. https://doi.org/10.3390/buildings10020024

Özeren, Ö., & Korumaz, M. (2021). Lidar to HBIM for analysis of historical buildings. Advanced LiDAR, 1(1), 27-
31. https://publish.mersin.edu.tr/index.php/lidar/article/view/58

Panoraven. (2021, September 01). The complete beginners guide to VR and 360 photography. Panoraven. Retrieved
April 26 from https://panoraven.com/blog/en/beginners-guide-360-photography/

Park, H., & Lee, D. (2019). Comparison between point cloud and mesh models using images from an unmanned
aerial vehicle. Measurement, 138, 461-466. https://doi.org/10.1016/j.measurement.2019.02.023

Patrucco, G., Rinaudo, F., & Spreafico, A. (2019). A new handheld scanner for 3D survey of small artifacts: The
Stonex F6. International Archives of the Photogrammetry, Remote Sensing and Spatial Information
Sciences - ISPRS Archives, 42(2/W15), 895-901. https://doi.org/10.5194/isprs-archives-XLII-2-W15-895-
2019

227

Pueschel, P. (2013). The influence of scanner parameters on the extraction of tree metrics from FARO Photon 120
terrestrial laser scans. ISPRS Journal of Photogrammetry and Remote Sensing, 78, 58-68.
https://doi.org/10.1016/j.isprsjprs.2013.01.005

Putch, A. (2022). Free SiteScape accounts & SiteScape pro. SiteScape. Retrieved April 26 from
https://support.sitescape.ai/hc/en-us/articles/4419890618132-Free-SiteScape-Accounts-SiteScape-Pro

Raychev, J., Hristov, G., Kyuchukova, D., & Zahariev, P. (2017, July 10-12). Workflow for development of a
virtual museum that will provide better way for learning the cultural heritage. 2017 16th International
Conference on Information Technology Based Higher Education and Training (ITHET), Ohrid, Macedonia.

RIEGL. (2022). RIEGL VZ-400i. RIEGL. http://www.riegl.com/uploads/tx_pxpriegldownloads/RIEGL_VZ-
400i_Datasheet_2022-09-27.pdf

Rocchini, C., Cignoni, P., Montani, C., Pingi, P., & Scopigno, R. (2001). A low cost 3D scanner based on structured
light. Computer Graphics Forum, 20(3), 299-308. https://doi.org/10.1111/1467-8659.00522

Rodríguez, R. F., Ponte, M., Bento, R., & Cardoso, R. (2022). Potential of mobile application based on structure
from motion (SfM) photogrammetry to monitor slope fast erosion by runoff water. Catena, 216, 106359.
https://doi.org/10.1016/j.catena.2022.106359

Romano, M., Antonelli, M., Palombo, M. R., Rossi, M. A., & Agostini, S. (2022). Drone testing for 3D
reconstruction of massive mounted skeletons in museums: the case of Mammuthus meridionalis (Nesti
1825) from Madonna della Strada (Scoppito, L’Aquila, Italy). Historical Biology, 34(7), 1305-1314.
https://doi.org/10.1080/08912963.2021.1975278

Salemi, G., Achilli, V., & Cervato, C. (2005, 24-25 February). Data acquisition for cultural heritage navigation:
integration of panoramic imaging, terrestrial laser scanning and anaglyphs. Proceedings of the ISPRS
Working Group, Berlin, Germany.

Sánchez-Jiménez, D., Buchón-Moragues, F., Bravo, J. M., & Sánchez-Pérez, J. V. (2019). Estimation of the
precision of a structured light system in oil paintings on canvas. Sensors, 19(22), 4966.
https://doi.org/10.3390/s19224966

Setyawan, A. A., Taftazani, M. I., Bahri, S., Noviana, E. D., & Faridatunnisa, M. (2022). Drone LiDAR application
for 3D city model. Journal of Applied Geospatial Information, 6(1), 572-576.
https://doi.org/10.30871/jagi.v6i1.3462

Shang, L., & Wang, C. (2022). Three-dimensional reconstruction and protection of mining heritage based on lidar
remote sensing and deep learning. Mobile Information Systems, 2022, 1-8.
https://doi.org/10.1155/2022/2412394

Shao, J., Zhang, W., Mellado, N., Grussenmeyer, P., Li, R., Chen, Y., Wan, P., Zhang, X., & Cai, S. (2019).
Automated markerless registration of point clouds from TLS and structured light scanner for heritage
documentation. Journal of Cultural Heritage, 35, 16-24. https://doi.org/10.1016/j.culher.2018.07.013

Shao, L., Yang, S., Fu, T., Lin, Y., Geng, H., Ai, D., Fan, J., Song, H., Zhang, T., & Yang, J. (2022). Augmented
reality calibration using feature triangulation iteration-based registration for surgical navigation. Computers
in Biology and Medicine, 148, 105826. https://doi.org/10.1016/j.compbiomed.2022.105826

She, M., Weiß, T., Song, Y., Urban, P., Greinert, J., & Köser, K. (2022). Marine bubble flow quantification using
wide-baseline stereo photogrammetry. ISPRS Journal of Photogrammetry and Remote Sensing, 190, 322-
341. https://doi.org/10.1016/j.isprsjprs.2022.06.014

Shet, A. (2022). Photogrammetry, 3D scanning with a photo camera. Bitfab. Retrieved April 26 from
https://bitfab.io/blog/photogrammetry/

228

Shu, F., Niu, S., Zhu, B., Wu, L., Xia, H., Chen, B., Zhao, J., & Tan, C. (2022). Influence of scan line spacing on
nanosecond pulse laser welding of 6063 Al to steel thin sheets. Optics & Laser Technology, 145, 107497.
https://doi.org/10.1016/j.optlastec.2021.107497

Shum, H. Y., & Szeliski, R. (1999, 20-27 September). Stereo reconstruction from multiperspective panoramas.
Proceedings of the Seventh IEEE International Conference on Computer Vision, Kerkyra, Greece.

Simek, J. F., Alvarez, S., & Cressler, A. (2022). Discovering ancient cave art using 3D photogrammetry: pre-contact
Native American mud glyphs from 19th Unnamed Cave, Alabama. Antiquity, 96(387), 662-678.
https://doi.org/10.15184/aqy.2022.24

Siwiec, J., & Lenda, G. (2022). Integration of terrestrial laser scanning and structure from motion for the assessment
of industrial chimney geometry. Measurement, 199, 111404.
https://doi.org/10.1016/j.measurement.2022.111404

Slaton, D. (2005, April). Preservation Brief 43: The preparation and use of historic structure reports. National
Parks Service. Retrieved April 26 from https://www.nps.gov/tps/how-to-preserve/briefs/43-historic-
structure-reports.htm

Smith, L. (2006). Uses of heritage. Routledge.

Sofia, S., Maetzke, F. G., Crescimanno, M., Coticchio, A., La Mela Veca, D. S., & Galati, A. (2022). The efficiency
of LiDAR HMLS scanning in monitoring forest structure parameters: implications for sustainable forest
management. EuroMed Journal of Business, 17(3), 350-373. https://doi.org/10.1108/EMJB-01-2022-0017

Spreafico, A., Chiabrando, F., Losè, L. T., & Tonolo, F. G. (2021). The ipad pro built-in lidar sensor: 3d rapid
mapping tests and quality assessment. The International Archives of Photogrammetry, Remote Sensing and
Spatial Information Sciences, 43(B1-2021), 63-69. https://doi.org/10.5194/isprs-archives-XLIII-B1-2021-
63-2021

St. Olaf College. (2022). Precision vs accuracy. St. Olaf College. Retrieved April 26 from
https://wp.stolaf.edu/it/gis-precision-accuracy/

Stanley, T. (2013). Assessment of the FARO 3D focus laser scanner for forest inventory [Bachelor, University of
Southern Queensland]. Southern Queensland.
https://core.ac.uk/download/pdf/19730268.pdfhttps://core.ac.uk/download/pdf/19730268.pdf

Stein, S. (2022, February 07). Lidar is one of the iPhone and iPad Pro's coolest tricks: Here's what else it can do.
CNET. Retrieved April 26 from https://www.cnet.com/tech/mobile/lidar-is-one-of-the-iphone-ipad-coolest-
tricks-its-only-getting-better/

Strunk, J. L., Bell, D. M., & Gregory, M. J. (2022). Pushbroom photogrammetric heights enhance state-level forest
attribute mapping with landsat and environmental gradients. Remote Sensing, 14(14), 3433.
https://doi.org/10.3390/rs14143433

Suchocki, C. (2020). Comparison of time-of-flight and phase-shift TLS intensity data for the diagnostics
measurements of buildings. Materials, 13(2), 353. https://doi.org/10.3390/ma13020353

Suchocki, C., Błaszczak-Bąk, W., Janicka, J., & Dumalski, A. (2021). Detection of defects in building walls using
modified OptD method for down-sampling of point clouds. Building Research & Information, 49(2), 197-
215. https://doi.org/10.1080/09613218.2020.1729687

Sully, D. (2008). Decolonizing conservation: caring for Maori meeting houses outside new Zealand. Routledge.

229

Taniguchi, K., Kubota, S., & Yasumuro, Y. (2022). Quantitative visualization of physical barriers for vulnerable
pedestrians based on photogrammetry. Construction Innovation, 22(3), 604-623.
https://doi.org/10.1108/CI-04-2021-0087

Tavani, S., Billi, A., Corradetti, A., Mercuri, M., Bosman, A., Cuffaro, M., Seers, T., & Carminati, E. (2022).
Smartphone assisted fieldwork: Towards the digital transition of geoscience fieldwork using LiDAR-
equipped iPhones. Earth-Science Reviews, 227, 103969. https://doi.org/10.1016/j.earscirev.2022.103969

Taylor, T. S. (2019). Introduction to laser science and engineering. CRC Press. https://doi.org/10.1201/b22159

TechTarget Contributor. (2016, March). 360-degree photograph. TechTarget. Retrieved April 26 from
https://www.techtarget.com/whatis/definition/360-degree-photograph

Ternon, Q., Danet, V., Thiriet, P., Ysnel, F., Feunteun, E., & Collin, A. (2022). Classification of underwater
photogrammetry data for temperate benthic rocky reef mapping. Estuarine, Coastal and Shelf Science, 270,
107833. https://doi.org/10.1016/j.ecss.2022.107833

Travis, C. M. J. (1994, September). Preservation brief 35 understanding old buildings: the process of architectural
investigation. National Historic Preservation. Retrieved April 26 from https://www.nps.gov/tps/how-to-
preserve/briefs/35-architectural-investigation.htm

Trimble. (2022). Trimble RealWorks. Trimble. Retrieved April 26 from https://geospatial.trimble.com/products-and-
solutions/trimble-realworks

UNESCO. World heritage. The United Nations Educational Scientific and Cultural Organization. Retrieved April 26
from https://whc.unesco.org/en/about/

Vavrouchová, H., Vaishar, A., & Peřinková, V. (2022). Historical landscape elements of abandoned foothill
villages—A case study of the historical territory of Moravia and Silesia. Land, 11(10), 1809.
https://doi.org/10.3390/land11101809

Vellanoweth, R. L., Porcayo-Michelini, A., Guttenberg, R. B., Hayden, W., Ainis, A. F., & Hernández-Estrada, R.
L. (2022). Spring tides, storm surges, and the destruction of coastal middens: A case study from the upper
Gulf of California, México. The Journal of Island and Coastal Archaeology, 17(3), 420-431.
https://doi.org/10.1080/15564894.2020.1782539

vGIS. (2020, December). iPhone and iPad LiDAR spatial tracking capabilities: second test. VGIS Inc. Retrieved
April 26 from https://www.vgis.io/2020/12/02/lidar-in-iphone-and-ipad-%20spatial-tracking-capabilities-
test-take-2/

Vizzo, I., Mersch, B., Marcuzzi, R., Wiesmann, L., Behley, J., & Stachniss, C. (2022). Make it dense: Self-
supervised geometric scan completion of sparse 3d lidar scans in large outdoor environments. IEEE
Robotics and Automation Letters, 7(3), 8534-8541. https://doi.org/10.1109/LRA.2022.3187255

Wachowiak, M. J., & Karas, B. V. (2009). 3D scanning and replication for museum and cultural heritage
applications. Journal of the American Institute for Conservation, 48(2), 141-158.
https://doi.org/10.1179/019713609804516992

Wang, M. (2022). Estimating vehicle speed with consumer grade mobile LiDAR [Doctoral dissertation,
Massachusetts Institute of Technology]. Massachusetts.
https://dspace.mit.edu/handle/1721.1/144817https://dspace.mit.edu/handle/1721.1/144817

White, D. A. (2023, March 25). Case study: Documentation of the Neutra Reunion House. Neutra Reunion House -
WCAPT documentation workshop, Los Angeles, CA.

230

Whitehead, D. A., Ayres, K. A., Gayford, J. H., Ketchum, J. T., Galván-Magana, F., & Christiansen, F. (2022).
Aerial photogrammetry of whale sharks (Rhincodon typus) in the Bay of La Paz, using an unoccupied
aerial vehicle. Marine Biology, 169(7), 94. https://doi.org/10.1007/s00227-022-04085-0

Xu, C., Li, Z., Li, H., Wang, F., & Zhou, P. (2019). Long-range terrestrial laser scanning measurements of annual
and intra-annual mass balances for Urumqi Glacier No. 1, eastern Tien Shan, China. The
Cryosphere, 13(9), 2361-2383. https://doi.org/10.5194/tc-13-2361-2019

Yastikli, N. (2007). Documentation of cultural heritage using digital photogrammetry and laser scanning. Journal of
Cultural heritage, 8(4), 423-427. https://doi.org/10.1016/j.culher.2007.06.003

Yoon, S. H. (2006, July 26-28). A surface displaced from a manifold. Geometric Modeling and Processing-GMP
2006: 4th International Conference, Pittsburgh, PA, USA.

Zhang, W., Gao, F., Jiang, N., Zhang, C., & Zhang, Y. (2022). High-temporal-resolution forest growth monitoring
based on segmented 3D canopy surface from UAV aerial photogrammetry. Drones, 6(7), 158.
https://doi.org/10.3390/drones6070158

Zhou, G., Li, C., Zhang, D., Liu, D., Zhou, X., & Zhan, J. (2021). Overview of underwater transmission
characteristics of oceanic LiDAR. IEEE Journal of Selected Topics in Applied Earth Observations and
Remote Sensing, 14, 8144-8159. https://doi.org/10.1109/JSTARS.2021.3100395

Zoller + Fröhlich. (2022). Laserscanning software. Zoller + Fröhlich. Retrieved April 16 from
https://www.zofre.de/en/laser-scanners/laserscanning-software

231

Appendix A

Data Acquisition for Test Scan Comparison
Three scans will be acquired by the same smartphone application SiteScape with low, medium, and high
point density. Three different point density scans were processed to examine if overlapping will help
increase points density and precision in describing scanning target.

Scan 1 (Low point density)
I. Preparation
a) Repeat 4.1.1 I. Preparation, set point density to Low and point size Low (Figures B-1, B-2)

Figures B-1 and B-2: Scan 1 point density parameters

II. Data Acquisition
a) Repeat 4.1.1 II. Data Acquisition with method A or B

232

III. Save and Export
a) Repeat 4.1.1 III. Save and Export as scan 1

Scan 2 (Medium point density)

I. Preparation
a) Repeat 4.1.1 I. Preparation, set point density to Med and point size Low (Figure B-3, B-4)

Figure B-3 andB-4: Scan 2 point density parameters

II. Data Acquisition
a) Repeat 4.1.1 II. Data Acquisition with method A or B

III. Save and Export

233

a) Repeat 4.1.1 III. Save and Export as scan 2

Scan 3 (High point density)
I. Preparation
a) Repeat 4.1.1 I. Preparation, set point density to High and point size Low (Figure B-5, B-6)

Figure B-5 and B-6: Scan 3 point density parameters

II. Data Acquisition
a) Repeat 4.1.1 II. Data Acquisition with method A or B

III. Save and Export
a) Repeat 4.1.1 III. Save and Export scan 3

234

Data Processing
The post-acquisition process includes synchronizing to cloud and downloading from SiteScape webpage.

With the raw scanning data, scan 1 (low point density) and scan 2 (medium point density) were selected
for following examination on the overlapping process proposed in Chapter 3. Scan 3 (high point density)
results in overly thick points on surfaces, which will impact alignment and comparison. To increasing the
points describing objects, a methodology of overlapping scan 1 and 2 into one 3D space was proposed.
Scan 1 and scan 2 were noise cleared and cropped as steps mentioned in 4.1.2.

Following the methodology proposed in Chapter 3, comparison set 1overlapped processed data of scan 1
and scan 2 together; comparison set 2 overlapped raw data of scan 1 and scan 2, then cleared noise and
cropped (Figure B-7).

Figure B-7: Documentation data process workflow

I. Download File
Data acquired from smartphone for the purpose of heritage conservation documentation should be
downloaded and ready for computer processing.
a) Download scanned data in PLY file format to computer; then open CloudCompare.

235

b) Click File-Open, change file format to PLY mesh, select the scan file, and open in CloudCompare.
By this step, the scanned data should be able to view in CloudCompare in point clouds (Figures 4-
77).

Figures B-8: Open file in CloudCompare

c) Observe and document the number of points from properties window for all three scans (Figures B-9,
B-10).

236

Figure B-9: Number of points for scan 1

Figure B-10: Number of points for scan 2

II. Data Preparation

237

To improve raw data’s operability in examining the ability of smartphone scanned data, scans should be
automatic noise cleaned and cropped.
a) Automatic noise clean and crop both scans 1 and 2 following the 4.1.1 instruction.
b) Observe and document the number of points from properties window for scan 1, scan 1 filter, scan 2,
scan 2 filter (Figures B-11, B-12).

Figure B-11: Number of points of scan 1 filter

238

Figure B-12: Number of points for scan 2 filter

c) Save scan 1 filter and scan 2 filter as individual files

3. Data Comparison
Processed scans are ready for comparison. The purpose of the comparisons is to understand how
overlapping impact smartphone scanned data’s quality.
a) Open a new CloudCompare window; import scan 1 and scan 2 in one file.
b) Automatic noise clean and cropping the combined file following instructions from 4.1.1
c) Save the file as scan 1+2 merge-filter-crop, observe and document the number of points from the
properties window (Figure B-13).

Figure B-13: Number of points of scan 1+2 merge-filter-crop

d) Open a new CloudCompare window, import scan 1 filter and scan 2 filter in the same file.

239

e) Save file as scan 1+2 filter-crop-merge, observe and document the number of points from properties
window (Figure B-14).

Figure B-14: Number of points of scan 1+2 merge-filter-crop

f) Compare the change in the number of points to see if overlapping point clouds would increase the
number describing the same target, as well as if the number of total point number is the sum of two
scans. If two points can be overlapped and the number of total point increase, more points are
describing the same surface.

The number of points as scanning results are different for different point densities (Table B-1)

240

Table B-1 SiteScape test scan number of points
Test Scan Number Parameters (Point Density) Scan Method Points Obtained
Scan 1 Low Handheld 3,550,490
Scan 2 Medium Handheld 5,768,038

The number of points in raw and processed files are smaller for each scan (Table B-2)

Table B-2 Results of data processing for HC documentation
Scans Number of Points Point Cleared
Scan 1 Filter 2,764,678
785,812 Scan 1 3,550,490
Scan 2 Filter 5,540,899
227,139 Scan 2 5,768,038
Scan 1+2 Merge 9,318,528

Scan 1+2 Filter-Crop-Merge 8,305,577 1,012,951
Scan 1+2 Merge-Filter-Crop 7,091,553 2,226,975

Results in table 4-2 reviewed that registering two or more point clouds together can increase the number
of points describing the object. The number of points for overlapped point clouds are the sum of the two
individual files. Scan 1 merge with scan 2 had 3,550,490 (scan 1) pulsed 5,768,038 (scan 2), resulted in
9,318,528 points in file scan 1+2. Scan 1+2 filter-crop-merge 8,305,577 points which was 5,540,899
(scan 1 filter) +2,764,678 (scan 2 filter). However, merging two point clouds can increase both the
number of points describing the object, and error points. It was partially proved by the difference in
number of points resulted from different process order that Scan 1+2 Merge-Filter-Crop had a smaller
number than Scan 1+2 Filter-Crop-Merge. The result showed filtering a combined file can take out more

241

fussy points than combining the two filtered files. These increased number on erased points indicate more
noise points can be identified than only add up the number of noise point from individual files. With more
noise cleared from the combined file, Scan 1+2 Merge-Filter-Crop would have increased precision and
more points describing the object. To further validating the accuracy of this method, more analysis will be
done in the case study.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

In-situ quality assessment of scan data for as-built models using building-specific geometric features

PDF

The ambivalence of conserving Busan’s colonial heritage

PDF

Mobilizing heritage conservation as a tool for urban resilience: linkages and recommendations

PDF

Reconstruction right now: conserving vernacular heritage in Beaufort, South Carolina as an act of reconstructing preservation practice

PDF

Deconstruction: a tool for sustainable conservation

PDF

Planning in advance for rehabilitation and restoration using BIM and seismic simulation to record and analyze the Japanese House in Los Angeles

PDF

Yuba-Sutter: a case study for heritage conservation in Punjabi-American communities

PDF

Mapping heritage justice in the city of Los Angeles

PDF

Managing change on Newbury Street: a case study analysis

PDF

Heritage in practice: a study of two urban rivers

PDF

Legacy business program implementation in American urban immigrant neighborhoods: a case study of Little Tokyo and Chinatown, Los Angeles

PDF

Identifying and conserving Pacoima: a heritage conservation study of a minority enclave in the San Fernando Valley

PDF

Empowering communities through historic rehabilitation: creating a maintenance plan for public housing developments in Los Angeles

PDF

Interpreting historic murals: a case study approach and best practices used in Fengguo Temple

PDF

The tent, camel, and coffee: safeguarding the intangible cultural heritage of the Rum Village Bedouins

PDF

Acoustic heritage of recording studios: physical characteristics and signature sound

PDF

Not so Little Armenia: conserving Armenian heritage sites in Los Angeles

PDF

Using the UNESCO Historic Urban Landscape framework: a case study of the Pico-Union neighborhood

PDF

Towards building a live 3D digital twin of the world

PDF

Would community design overlay (CDO) be a tool for revitalizing Los Angeles Chinatown?

Asset Metadata

Creator Hong, Ye (author)

Core Title Scanning using a smart phone for heritage conservation: a case study using the Reunion House

School School of Architecture

Degree Master of Building Science and Heritage Conservation

Degree Program Building Science / Heritage Conservation

Degree Conferral Date 2023-05

Publication Date 05/08/2023

Defense Date 03/29/2023

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag 3D scanning,handheld mobile laser scanning,Heritage Conservation,LiDAR,OAI-PMH Harvest,point clouds,smartphone

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Kensek, Karen (committee chair), Hall, Peyton (committee member), Sandmeier, Trudi (committee member)

Creator Email cchy9297@outlook.com,yehong@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113102884

Unique identifier UC113102884

Identifier etd-HongYe-11797.pdf (filename)

Legacy Identifier etd-HongYe-11797

Document Type Thesis

Format theses (aat)

Rights Hong, Ye

Internet Media Type application/pdf

Type texts

Source 20230508-usctheses-batch-1039 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu