TY - JOUR
T1 - Pixel-level multicategory detection of visible seismic damage of reinforced concrete components
AU - Miao, Zenghui
AU - Ji, Xiaodong
AU - Okazaki, Taichiro
AU - Takahashi, Noriyuki
N1 - Funding Information:
The work presented in this paper was sponsored by the funds from the National Key Research and Development Program of China (Grant No. 2017YFC1500602), the NSFC Project of International Cooperation and Exchanges (NSFC‐JSPS) (Grant No. 51811540032), the Tsinghua University Initiative Scientific Research Program (Grant No. 20193080019), and the National Natural Science Foundation of China (Grant No. 52078277). The financial support is sincerely appreciated.
Publisher Copyright:
© 2021 Computer-Aided Civil and Infrastructure Engineering
PY - 2021/5
Y1 - 2021/5
N2 - The detection of visible damage (i.e., cracking, concrete spalling and crushing, reinforcement exposure, buckling and fracture) plays a key role in postearthquake safety assessment of reinforced concrete (RC) building structures. In this study, a novel approach based on computer-vision techniques was developed for pixel-level multicategory detection of visible seismic damage of RC components. A semantic segmentation database was constructed from test photos of RC structural components. Series of datasets were generated from the constructed database by applying image transformations and data-balancing techniques at the sample and pixel levels. A deep convolutional network architecture was designed for pixel-level detection of visible damage. Two sets of parameters were optimized separately, one to detect cracks and the other to detect all other types of damage. A postprocessing technique for crack detection was developed to refine crack boundaries, and thus improve the accuracy of crack characterization. Finally, the proposed vision-based approach was applied to test photos of a beam-to-wall joint specimen. The results demonstrate the accuracy of the vision-based approach to detect damage, and its high potential to estimate seismic damage states of RC components.
AB - The detection of visible damage (i.e., cracking, concrete spalling and crushing, reinforcement exposure, buckling and fracture) plays a key role in postearthquake safety assessment of reinforced concrete (RC) building structures. In this study, a novel approach based on computer-vision techniques was developed for pixel-level multicategory detection of visible seismic damage of RC components. A semantic segmentation database was constructed from test photos of RC structural components. Series of datasets were generated from the constructed database by applying image transformations and data-balancing techniques at the sample and pixel levels. A deep convolutional network architecture was designed for pixel-level detection of visible damage. Two sets of parameters were optimized separately, one to detect cracks and the other to detect all other types of damage. A postprocessing technique for crack detection was developed to refine crack boundaries, and thus improve the accuracy of crack characterization. Finally, the proposed vision-based approach was applied to test photos of a beam-to-wall joint specimen. The results demonstrate the accuracy of the vision-based approach to detect damage, and its high potential to estimate seismic damage states of RC components.
UR - http://www.scopus.com/inward/record.url?scp=85101474433&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85101474433&partnerID=8YFLogxK
U2 - 10.1111/mice.12667
DO - 10.1111/mice.12667
M3 - Article
AN - SCOPUS:85101474433
SN - 1093-9687
VL - 36
SP - 620
EP - 637
JO - Computer-Aided Civil and Infrastructure Engineering
JF - Computer-Aided Civil and Infrastructure Engineering
IS - 5
ER -