Built Environment Project and Asset Management
Vision a
nd advocacy of optoelectronic technology
developments in the AECO sector
Journal:
Built Environment Project and Asset Management
Manuscript ID
BEPAM-11-2016-0081.R2
Manuscript Type:
Review Paper
Keywords:
building information modelling (BIM), laser scanning, machine vision,
digital built environment, optoelectronic devices, point cloud
Built Environment Project and Asset Management
Built Environment Project and Asset Management
1
Vision and advocacy of optoelectronic technology developments in the
AECO sector
ABSTRACT
Purpose: This research presents a literature review of laser scanning and 3D modelling
devices, modes of delivery and applications within the architecture, engineering, construction
and owner-operated (AECO) sector. Such devices are inextricably linked to modern digital
built environment practices, particularly when used in conjunction with as-built building
information modelling (BIM) development. The research also reports upon innovative
technological advancements (such as machine vision) that coalesce with 3D scanning
solutions.
Design: A synthesis of literature is used to develop: a hierarchy of the modes of delivery for
laser scan devices; a thematic analysis of 3D terrestrial laser scan technology applications;
and a componential cross-comparative tabulation of laser scan technology and specifications.
Findings: Findings reveal that the costly and labour intensive attributes of laser scanning
devices have stimulated the development of hybrid automated and intelligent technologies to
improve performance. Such developments are set to satisfy the increasing demand for
digitisation of both existing and new buildings into BIM. Future work proposed will seek to:
review what coalescence of digital technologies will provide an optimal and cost effective
solution to accurately reconstructing the digital built environment; conduct case studies that
implement hybrid digital solutions in pragmatic facilities management scenarios to measure
their performance and user satisfaction; and eliminate manual remodelling tasks (such as
point cloud reconstruction) via the use of computational intelligence algorithms integral
within cloud based BIM platforms.
Originality: Although laser scanning and 3D modelling have been widely covered en
passant within the literature, scant research has conducted an holistic review of the
technology, its applications and future developments. This review presents concise and lucid
reference guidance that will intellectually challenge, and better inform, both practitioners and
researchers.
Page 1 of 28 Built Environment Project and Asset Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Built Environment Project and Asset Management
2
KEYWORDS
Laser scanning, digital built environment, building information modelling (BIM), machine
vision
INTRODUCTION
Optoelectronics has its etymological roots grounded in physics and encompasses the study,
design and manufacture of electronic devices for emitting, modulating, transmitting and
sensing light (Sergiyenko and Rodriguez-Quiñonez, 2016). Optoelectronic devices gather and
display information at high speed but can also store and process information (Marzuki, 2016).
Beneficial characteristics of these devices include their: small and portable size (Faro, 2004);
highly sophisticated functionality (Marzuki, 2016); solid-state robustness (Lindner, 2016);
and low-power consumption during operation (Faro, 2004). Such attributes have engendered
the proliferation of optoelectronic devices throughout society, business and commerce
(Rushmeier, 2002). Amongst the hierarchy of optoelectronic technologies available, optical
laser scanners are frequently used for the rapid automation of millimetre precision
measurement and reconstruction of tangible objects via processed optical signals from
reflected light (Thiel and Wehr, 2004). Laser scanning applications are myriad throughout a
disparate range of industries, including: aerospace for structural health monitoring (Derriso et
al., 2016); law enforcement for virtual crime scene reconstruction (Buck et al., 2013);
agriculture for crop growth monitoring (Cointault et al., 2016); archaeology for re-
constructing archaeological artefacts (Galeazzi et al., 2016); and manufacturing industries for
quality assurance purposes (Godin et al., 1994; Mello and Stemmer, 2016). Given this strong
demand, the laser scanning industry’s value is forecast to exceed 5.90 Billion USD by 2022
(Markets, 2016).
The architecture, engineering, construction and owner-operated (AECO) sector encompasses
the whole life cycle of buildings and infrastructure within the built environment. Within this
sector 3D laser distance and ranging (LiDAR) devices rapidly construct point cloud data sets
that precisely measure large volumes of physical objects, transcribed into a digital built
environment (Chen et al., 2015). Laser scan devices are now integral within numerous built
environment applications including: construction progress tracking (El-Omari and Moselhi,
2008); quality control assessment (Wang et al., 2016); site activity monitoring (Zhang et al.,
2016); safety assessment (Shapira et al., 2014); and resource and material tracking (Szweda,
2006). Laser scanning also represents an ideal technological solution for automating as-built
Page 2 of 28Built Environment Project and Asset Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Built Environment Project and Asset Management
3
Building Information Modelling (BIM) validation (and model updates) for contractors and
facility managers (Hoffmeister, 2016). However, whilst demand has grown, laser scan
technology is not universally adopted in contemporary AECO practice because of issues
pertaining to: high equipment costs (Rushmeier, 2002; Qin and Gruen, 2014); costly human
resource training needed to acquire core competencies to operate the laser scanning
equipment (Ouédraogo et al., 2014); discrepancies in spatial information ( Tang and Akinci,
2012a, 2012b; Jung et al., 2014;
Laing et al., 2015) lack of automation with BIM object
recognition from point cloud data (Ouédraogo et al., 2014); timely calibration of scanning
equipment (Kim et al., 2015a); limited point cloud capture from occlusions (Xiong et al.,
2013); and excessive time consumed with point cloud data processing (Golparvar-Fard et al.,
2011). In addition, laser-scanning applications integrated with BIM lack automation in the
recognition of semantic attributes from scanned data – albeit point cloud semantic recognition
has accrued maturity in factory automated scanning of small objects (Rushmeier, 2002;
Godin et al., 2010).
To address these limitations, research and industry practice advocate using automated image-
based sensing technologies to augment laser scan technology (Golparvar-Fard et al., 2011).
Machine vision enhances laser scan capabilities through object recognition and description,
utilising algorithms for image processing and pattern recognition (Gao et al., 2015). Such
advancements have been successfully applied within the manufacturing, automotive and
security industries – hence, advanced technology transfer into the AECO sector is feasible
(Flores-Fuentes et al., 2016). The wider applications of integrated scanning-software devices
represent the next generation of revolutionary technological solutions adopted throughout a
building’s entire life cycle (Turkan et al., 2014). Hitherto, research has not reported upon
suitable automated image/laser-based reconstruction to manage and update the as-built BIM
for both new build development and/or retrospective modelling. Against this contextual
backdrop, this literature review aims to: i) provide a qualitative, componential analysis of
current as-built image/laser-based reconstruction devices and applications used to maintain
digital built environment data; ii) critically evaluate contemporary laser scanning and
automated machine vision processing developments; and iii) propose a future trajectory for
augmented laser scanning devices in the AECO sector.
Page 3 of 28 Built Environment Project and Asset Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Built Environment Project and Asset Management
4
FUNDAMENTALS OF LASER-SCAN DEVICES
Optoelectronic scanning devices have developed rapidly since initial laboratory testing was
conducted by the National Research Council of Canada (NRCC) during the 1970s (Blais,
2004) (refer to Figure 1). 3D laser scanning initially lagged behind 2D laser scanning
technologies in terms of image quality, rendering and ease of use (ibid). However, the advent
of microcomputers and exponential improvements in computational processing power
transformed 3D laser scanning to mirror the performance of its counterpart (Sergiyenko and
Rodriguez-Quiñonez, 2016). These technological breakthroughs have stimulated the
development of cost-effective, automated, contactless 3D range sensor systems for various
industrial applications (Hornberg, 2006). Since the early 1980s, three generic contactless
methods of range measurement have emerged, namely: i) triangulation; ii) phase shift; and
iii) time of flight (TOF) or pulse-based – albeit, a number of hybrid scanning devices
integrate multiple methods (e.g Moiré pattern projection and fringe interference) (Creath and
Wyant, 1992).
Triangulation
Triangulation is suitable for short-to-medium distance measurements (<5-10 m). An early
example of triangulation based 3D laser cameras are slit scanners due to their optical and
mechanical simplicity and cost (Galantucci et al., 2015). Slit-scanners utilise an angled light
line projected onto the target object; deformation of the straight-line projection provides
information about the target surface protuberances (Hornberg, 2006). This method is limited
in capacity for large volume capture since short-range slit scanners are a compromise
between field of view and depth resolution (ibid). They are typically limited to a field of view
of between 20-50 degrees.
Phase shift
Phase shift devices calculate the difference between the overlapping sent and reflected
signal(s) within a certain wavelength (Lindner, 2016). This offers speedy measurement in
comparison with TOF methods but the output point cloud data does suffer from higher
speckle noise interference (Creath and Wyant, 1992; Beraldin et al., 2000a). Whilst speckle
noise can be reduced with good lighting conditions, such instances are rare on construction
sites, particularly when laser scanning larger objects such as buildings (Godin et al., 2010).
Page 4 of 28Built Environment Project and Asset Management
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60