TY - JOUR
T1 - Synthetic environments for vision-based structural condition assessment of Japanese high-speed railway viaducts
AU - Narazaki, Yasutaka
AU - Hoskere, Vedhus
AU - Yoshida, Koji
AU - Spencer, Billie F.
AU - Fujino, Yozo
N1 - Funding Information:
The authors would like to acknowledge the financial support by the U.S. Army Corps of Engineers (Contract/Purchase Order No. W912HZ-17-2-0024). This research was also supported in part by the National Natural Science Foundation of China Grant No. 51978182. The authors would like to acknowledge Professor Yoshikazu Takahashi at the Kyoto University for providing images of damaged RC railway viaducts that were needed for this research.
Publisher Copyright:
© 2021 Elsevier Ltd
PY - 2021/11
Y1 - 2021/11
N2 - Civil infrastructure condition assessment using visual recognition methods has shown significant potential for automating various aspects of the problem, including identification and localization of critical structural components, as well as detection and quantification of structural damage. The application of those methods typically requires large amounts of training data that consists of images and corresponding ground truth annotations. However, obtaining such datasets is challenging, because the images are annotated manually in most existing approaches. With the limited availability of datasets, development of effective visual recognition systems that can extract all required information is not straightforward. This research leverages synthetic environments to develop a unified system for automated vision-based structural condition assessment that can identify and localize critical structural components, and then detect and quantify damage of those components. The synthetic environments can produce images and associated ground truth annotations for semantic segmentation of structural components and damage, as well as monocular depth estimation for structural component localization. To illustrate the approach, automated vision-based structural condition assessment of reinforced concrete railway viaducts for a Japanese high-speed railway line (the Tokaido Shinkansen) is explored. The effectiveness of the synthetic environments and the generated dataset (the Tokaido dataset) is demonstrated by training fully convolutional network-based semantic segmentation and monocular depth estimation algorithms, and then testing the networks using both synthetic and real-world images. Finally, all trained algorithms are combined to realize an automated system for structural condition assessment.
AB - Civil infrastructure condition assessment using visual recognition methods has shown significant potential for automating various aspects of the problem, including identification and localization of critical structural components, as well as detection and quantification of structural damage. The application of those methods typically requires large amounts of training data that consists of images and corresponding ground truth annotations. However, obtaining such datasets is challenging, because the images are annotated manually in most existing approaches. With the limited availability of datasets, development of effective visual recognition systems that can extract all required information is not straightforward. This research leverages synthetic environments to develop a unified system for automated vision-based structural condition assessment that can identify and localize critical structural components, and then detect and quantify damage of those components. The synthetic environments can produce images and associated ground truth annotations for semantic segmentation of structural components and damage, as well as monocular depth estimation for structural component localization. To illustrate the approach, automated vision-based structural condition assessment of reinforced concrete railway viaducts for a Japanese high-speed railway line (the Tokaido Shinkansen) is explored. The effectiveness of the synthetic environments and the generated dataset (the Tokaido dataset) is demonstrated by training fully convolutional network-based semantic segmentation and monocular depth estimation algorithms, and then testing the networks using both synthetic and real-world images. Finally, all trained algorithms are combined to realize an automated system for structural condition assessment.
KW - Automated structural inspection
KW - Monocular depth estimation
KW - Railway viaduct
KW - Reinforced concrete
KW - Semantic segmentation
KW - Synthetic environment
UR - http://www.scopus.com/inward/record.url?scp=85104354450&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85104354450&partnerID=8YFLogxK
U2 - 10.1016/j.ymssp.2021.107850
DO - 10.1016/j.ymssp.2021.107850
M3 - Article
AN - SCOPUS:85104354450
SN - 0888-3270
VL - 160
JO - Mechanical Systems and Signal Processing
JF - Mechanical Systems and Signal Processing
M1 - 107850
ER -