Computer Vision and Pattern Recognition Technologies for Construction - Mobile and Pervasive Computing in Construction - page 199

10.3.6 Global building damage
Accurate evaluation of damage sustained by buildings during catastrophic events,
for example, earthquakes or terrorist attacks, is critical to determine the buildings'
safety and their suitability for future occupancy. Rathje and Crawford (2004)
present a method that uses photogrammetry and high resolution satellite images to
identify damage resulting from an earthquake. Kamat and El-Tawil (2007) report
the results of experiments that were conducted to test that previously stored
building information (CAD images) can be superimposed onto a real structure in
Augmented Reality, and that structural damage can be detected by measuring and
interpreting key differences between a baseline image and the real view of the
facility. In addition to the experimental results, they designed a method to compute
a global building damage measure (interstory drift ratio - IDR).
10.3.7 Pavement surface
In the past two decades a huge amount of research has been carried out in
automating pavement distress detection, classification, assessment and repair.
The most popular approaches are based computer vision algorithms that operate
on 2D images to recognize, classify and measure pavement surface defects. These
approaches have been developed with specific regard to crack detection and
assessment, in particular real-time crack analysis (Wang and Gong, 2005;
Huang and Xu, 2006), crack classification into transverse, longitudinal and alligator
types (Sun et al ., 2009), crack depth estimation (Amarasiri et al ., 2010), and even
automating crack sealing (Haas, 1996; Kim et al ., 2009). Although many research
studies have addressed this problem and a few commercial software packages exist
(Fugro Roadware, 2004), fully automating pavement distress detection and clas-
sification in a real-time environment has remained a challenge due to varying
lighting conditions, shadows, and crack positions (Tsai et al ., 2010). Besides cracks,
vision-based approaches have also been applied to the detection of other pavement
defects, like potholes, patching, joints and raveling. Koch and Brilakis (2010)
present a pothole detection method combining shape and texture extraction
techniques resulting in a detection accuracy of 86%. Zhou et al . (2006) propose
a wavelet transform based method supporting real-time pavement distress detec-
tion, isolation, and evaluation. In Nguyen et al . (2009) an approach to detection of
joint and bridged defects using a measure of conditional texture anisotropy (CTA)
is presented. Furthermore, adaptive imaging techniques have been successfully
applied to detect cracks and patches, and measure defect areas (Cafiso et al ., 2006;
Battiato et al ., 2007).
10.4 Videogrammetric surveying
Two types of remote sensing technologies are commonly used today for construc-
tion site surveying: time-of-flight and visual triangulation. Time-of-flight based
sensors, such as terrestrial laser scanners and 3D range video cameras, work by