Structural damage-causing concrete cracking detection based on a deep-learning method

被引:28
作者
Han, Xiaojian [1 ]
Zhao, Zhicheng [1 ]
Chen, Lingkun [2 ,3 ,4 ,7 ]
Hu, Xiaolun [5 ]
Tian, Yuan [6 ]
Zhai, Chencheng [2 ]
Wang, Lu [2 ]
Huang, Xiaoming [5 ]
机构
[1] Nanjing Tech Univ, Coll Civil Engn, Nanjing 211800, Jiangsu, Peoples R China
[2] Yangzhou Univ, Coll Architecture Sci & Engn, Yangzhou 225127, Jiangsu, Peoples R China
[3] Univ Calif Los Angeles, Dept Civil & Environm Engn, Los Angeles, CA 90095 USA
[4] Southwest Jiaotong Univ, Sch Civil Engn, Chengdu 610031, Sichuan, Peoples R China
[5] Southeast Univ, Sch Transportat, Nanjing 211189, Jiangsu, Peoples R China
[6] China Acad Transportat Sci, Transportat Technol Dev Promot Ctr, Beijing 100029, Peoples R China
[7] Yangzhou Univ, Coll Civil Sci & Engn, 196 West Huayang Rd, Yangzhou 225127, Jiangsu, Peoples R China
关键词
Crack Recognition; Deep-learning; Transfer Learning; Deep convolutional neural networks (DCNNs); Digital image processing; Local threshold segmentation; CONVOLUTIONAL NEURAL-NETWORKS; SYSTEM;
D O I
10.1016/j.conbuildmat.2022.127562
中图分类号
TU [建筑科学];
学科分类号
0813 ;
摘要
Cracking is one of the common manifestations of damage in concrete structures. Crack detection is currently done using typical computer vision methods. But they are still not clever enough and have limited precision. Deep convolutional neural networks (DCNNs) are sophisticated models that recently improved visual task performance. This article proposes a hybrid technique based on CNN and digital image processing to identify fractures in photos. Using transfer learning, an AlexNet-based CNN is effectively trained on a tiny dataset and achieves 98.26% accuracy on fresh test data. Then the segments are cracked using a threshold-based technique. In the end, the crack mask is made up of all segments' shows. With transfer learning, the quantity of data and expenses necessary are lowered while retaining accuracy. Precise crack cover detection aids in automated crack measuring and fracture type labeling.
引用
收藏
页数:8
相关论文
共 37 条
  • [31] Development of an automatic detector of cracks in concrete using machine learning
    Yokoyama, Suguru
    Matsumoto, Takashi
    [J]. 3RD INTERNATIONAL CONFERENCE ON SUSTAINABLE CIVIL ENGINEERING STRUCTURES AND CONSTRUCTION MATERIALS - SUSTAINABLE STRUCTURES FOR FUTURE GENERATIONS, 2017, 171 : 1250 - 1255
  • [32] Yosinski J, 2014, ADV NEUR IN, V27
  • [33] Efficient Crack Detection Method for Tunnel Lining Surface Cracks Based on Infrared Images
    Yu, Tiantang
    Zhu, Aixi
    Chen, Yingying
    [J]. JOURNAL OF COMPUTING IN CIVIL ENGINEERING, 2017, 31 (03)
  • [34] Image Based Techniques for Crack Detection, Classification and Quantification in Asphalt Pavement: A Review
    Zakeri, H.
    Moghadas Nejad, Fereidoon
    Fahimifar, Ahmad
    [J]. ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING, 2017, 24 (04) : 935 - 977
  • [35] Anisotropic clustering on surfaces for crack extraction
    Zhao, Guoteng
    Wang, Tongqing
    Ye, Junyong
    [J]. MACHINE VISION AND APPLICATIONS, 2015, 26 (05) : 675 - 688
  • [36] A Comprehensive Survey on Transfer Learning
    Zhuang, Fuzhen
    Qi, Zhiyuan
    Duan, Keyu
    Xi, Dongbo
    Zhu, Yongchun
    Zhu, Hengshu
    Xiong, Hui
    He, Qing
    [J]. PROCEEDINGS OF THE IEEE, 2021, 109 (01) : 43 - 76
  • [37] Multicategory damage detection and safety assessment of post-earthquake reinforced concrete structures using deep learning
    Zou, Dujian
    Zhang, Ming
    Bai, Zhilin
    Liu, Tiejun
    Zhou, Ao
    Wang, Xi
    Cui, Wei
    Zhang, Shaodong
    [J]. COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING, 2022, 37 (09) : 1188 - 1204