Crack detection is a vital task to maintain a bridge's health and safety condition. Traditional computer-vision based methods easily suffer from disturbance of noise and clutters for a real bridge inspection. To address this limitation, we propose a two-stage crack detection approach based on Convolutional Neural Networks (CNN) in this letter. A predictor of small receptive field is exploited in the first detection stage, while another predictor of large receptive field is used to refine the detection results in the second stage. Benefiting from data fusion of confidence maps produced by both predictors, our method can predict the probability belongs to cracked areas of each pixel accurately. Experimental results show that the proposed method is superior to an up-to-date method on real concrete surface images.
Yundong LI
North China University of Technology
Weigang ZHAO
Shijiazhuang Tiedao University
Xueyan ZHANG
North China University of Technology
Qichen ZHOU
North China University of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Yundong LI, Weigang ZHAO, Xueyan ZHANG, Qichen ZHOU, "A Two-Stage Crack Detection Method for Concrete Bridges Using Convolutional Neural Networks" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 12, pp. 3249-3252, December 2018, doi: 10.1587/transinf.2018EDL8150.
Abstract: Crack detection is a vital task to maintain a bridge's health and safety condition. Traditional computer-vision based methods easily suffer from disturbance of noise and clutters for a real bridge inspection. To address this limitation, we propose a two-stage crack detection approach based on Convolutional Neural Networks (CNN) in this letter. A predictor of small receptive field is exploited in the first detection stage, while another predictor of large receptive field is used to refine the detection results in the second stage. Benefiting from data fusion of confidence maps produced by both predictors, our method can predict the probability belongs to cracked areas of each pixel accurately. Experimental results show that the proposed method is superior to an up-to-date method on real concrete surface images.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8150/_p
Copy
@ARTICLE{e101-d_12_3249,
author={Yundong LI, Weigang ZHAO, Xueyan ZHANG, Qichen ZHOU, },
journal={IEICE TRANSACTIONS on Information},
title={A Two-Stage Crack Detection Method for Concrete Bridges Using Convolutional Neural Networks},
year={2018},
volume={E101-D},
number={12},
pages={3249-3252},
abstract={Crack detection is a vital task to maintain a bridge's health and safety condition. Traditional computer-vision based methods easily suffer from disturbance of noise and clutters for a real bridge inspection. To address this limitation, we propose a two-stage crack detection approach based on Convolutional Neural Networks (CNN) in this letter. A predictor of small receptive field is exploited in the first detection stage, while another predictor of large receptive field is used to refine the detection results in the second stage. Benefiting from data fusion of confidence maps produced by both predictors, our method can predict the probability belongs to cracked areas of each pixel accurately. Experimental results show that the proposed method is superior to an up-to-date method on real concrete surface images.},
keywords={},
doi={10.1587/transinf.2018EDL8150},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - A Two-Stage Crack Detection Method for Concrete Bridges Using Convolutional Neural Networks
T2 - IEICE TRANSACTIONS on Information
SP - 3249
EP - 3252
AU - Yundong LI
AU - Weigang ZHAO
AU - Xueyan ZHANG
AU - Qichen ZHOU
PY - 2018
DO - 10.1587/transinf.2018EDL8150
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2018
AB - Crack detection is a vital task to maintain a bridge's health and safety condition. Traditional computer-vision based methods easily suffer from disturbance of noise and clutters for a real bridge inspection. To address this limitation, we propose a two-stage crack detection approach based on Convolutional Neural Networks (CNN) in this letter. A predictor of small receptive field is exploited in the first detection stage, while another predictor of large receptive field is used to refine the detection results in the second stage. Benefiting from data fusion of confidence maps produced by both predictors, our method can predict the probability belongs to cracked areas of each pixel accurately. Experimental results show that the proposed method is superior to an up-to-date method on real concrete surface images.
ER -