Automatic concrete infrastructure crack semantic segmentation using deep learning

被引:48
作者
Chen, Bo [1 ]
Zhang, Hua [1 ]
Wang, Guijin [2 ]
Huo, Jianwen [1 ]
Li, Yonglong [2 ]
Li, Linjing [1 ]
机构
[1] Southwest Univ Sci & Technol, Sch Informat Engn, Mianyang 621000, Sichuan, Peoples R China
[2] Tsinghua Univ, Dept Elect, Beijing 100084, Peoples R China
基金
中国国家自然科学基金;
关键词
Crack detection; Deep learning; Semantic segmentation; Statistical analysis; DAMAGE DETECTION; CIVIL INFRASTRUCTURE; COMPUTER VISION; INSPECTION; TEXTURE; NETWORK;
D O I
10.1016/j.autcon.2023.104950
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
To solve the privation of crack information during feature transmission, we propose an automatic concrete infrastructure crack semantic segmentation method using deep learning. Initially, based on the statistical analysis results of crack images, a multi-stage feature extraction network is created with multi-resolution parallel transmission to extract the crack features. Then, the extracted features are aggregated according to the corre-lation between segmentation classes and pixels to enhance the localization performance of the model for cracks. Moreover, using statistical analysis results as constraints construct the loss function to optimize the model and overcome the data imbalance issue. Experiments are conducted on a self-made manual annotation dataset, which contains 2000 images from the dam, the bridge, and the spillway tunnel, and our method reach 94.51% Pre-cision, 86.39% Recall, 82.26% Intersection-over-Unions, and 90.27% F1_measure on the dataset. The experi-mental results show that the proposed method is optimal for the semantic segmentation of crack images.
引用
收藏
页数:13
相关论文
共 42 条
[1]   Encoder-decoder network for pixel-level road crack detection in black-box images [J].
Bang, Seongdeok ;
Park, Somin ;
Kim, Hongjo ;
Kim, Hyoungkwan .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2019, 34 (08) :713-727
[3]   Autonomous concrete crack detection using deep fully convolutional neural network [J].
Cao Vu Dung ;
Le Duc Anh .
AUTOMATION IN CONSTRUCTION, 2019, 99 :52-58
[4]   Deep Learning-Based Crack Damage Detection Using Convolutional Neural Networks [J].
Cha, Young-Jin ;
Choi, Wooram ;
Buyukozturk, Oral .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2017, 32 (05) :361-378
[5]   Quantify pixel-level detection of dam surface crack using deep learning [J].
Chen, Bo ;
Zhang, Hua ;
Li, Yonglong ;
Wang, Shuang ;
Zhou, Huaifang ;
Lin, Haitao .
MEASUREMENT SCIENCE AND TECHNOLOGY, 2022, 33 (06)
[6]   A novel U-shaped encoder-decoder network with attention mechanism for detection and evaluation of road cracks at pixel level [J].
Chen, Jun ;
He, Ye .
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2022, 37 (13) :1721-1736
[7]   Machine vision-based model for spalling detection and quantification in subway networks [J].
Dawood, Thikra ;
Zhu, Zhenhua ;
Zayed, Tarek .
AUTOMATION IN CONSTRUCTION, 2017, 81 :149-160
[8]   A review of computer vision-based structural health monitoring at local and global levels [J].
Dong, Chuan-Zhi ;
Catbas, F. Necati .
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2021, 20 (02) :692-743
[9]   Efficient real-time defect detection for spillway tunnel using deep learning [J].
Feng, Chuncheng ;
Zhang, Hua ;
Li, Yonglong ;
Wang, Shuang ;
Wang, Haoran .
JOURNAL OF REAL-TIME IMAGE PROCESSING, 2021, 18 (06) :2377-2387
[10]   Structural Damage Detection using Deep Convolutional Neural Network and Transfer Learning [J].
Feng, Chuncheng ;
Zhang, Hua ;
Wang, Shuang ;
Li, Yonglong ;
Wang, Haoran ;
Yan, Fei .
KSCE JOURNAL OF CIVIL ENGINEERING, 2019, 23 (10) :4493-4502