摘要A concrete crack detection method suitable for various scene conditions is proposed on the basis of image recognition to complete the classification and the identification of the cracks of the collected crack images in the bridge health monitoring work conveniently and reliably. Moreover, this method can improve the crack recognition effect, which is greatly affected by the selection of the initial clustering center of the extraction algorithm, and has high environmental dependence on the image background. Convolutional neural networks (CNNs) is a representative deep learning algorithm that can characterize learning and classify input information according to its own hierarchical structure. After collecting image data on the spot by relying on the Baoxie River Bridge inspection project, Gaoxin 4th Road, Donghu High-tech Zone, Wuhan City based on the CNN, an image classification model suitable for concrete crack image classification is established. This model realizes the collection of the inspection project image classification of concrete structures in complex scenes while considering that the local cluster density and Euclidean distance of the clustering center in the traditional K-means algorithm are both large. The traditional K-means algorithm is improved by combining the use of statistical principles and morphological methods. Finally, the improved K-means algorithm completes the crack skeleton segmentation extraction and crack width calculation of crack images in complex scenes. The effectiveness of the proposed method under concrete surface peelings, stains, mosses, or other noise conditions was verified according to the successful identification of 600 crack on-site images photographed from a bridge surface. Results also show that the efficiency of the proposed crack detection method is higher than that of traditional methods. The proposed approach can also provide a reference for the in depth research on crack identification on the surface of concrete structures in complex background in the future.
Abstract：A concrete crack detection method suitable for various scene conditions is proposed on the basis of image recognition to complete the classification and the identification of the cracks of the collected crack images in the bridge health monitoring work conveniently and reliably. Moreover, this method can improve the crack recognition effect, which is greatly affected by the selection of the initial clustering center of the extraction algorithm, and has high environmental dependence on the image background. Convolutional neural networks (CNNs) is a representative deep learning algorithm that can characterize learning and classify input information according to its own hierarchical structure. After collecting image data on the spot by relying on the Baoxie River Bridge inspection project, Gaoxin 4th Road, Donghu High-tech Zone, Wuhan City based on the CNN, an image classification model suitable for concrete crack image classification is established. This model realizes the collection of the inspection project image classification of concrete structures in complex scenes while considering that the local cluster density and Euclidean distance of the clustering center in the traditional K-means algorithm are both large. The traditional K-means algorithm is improved by combining the use of statistical principles and morphological methods. Finally, the improved K-means algorithm completes the crack skeleton segmentation extraction and crack width calculation of crack images in complex scenes. The effectiveness of the proposed method under concrete surface peelings, stains, mosses, or other noise conditions was verified according to the successful identification of 600 crack on-site images photographed from a bridge surface. Results also show that the efficiency of the proposed crack detection method is higher than that of traditional methods. The proposed approach can also provide a reference for the in depth research on crack identification on the surface of concrete structures in complex background in the future.
基金资助:Supported by the Natural Science Foundation of Hubei Province (No. 2018CFB609)
通讯作者:
LEI Si-da
E-mail: 547586099@qq.com
引用本文:
雷斯达, 曹鸿猷, 康俊涛. Concrete Surface Crack Recognition in Complex Scenario Based on Deep Learning[J]. Journal of Highway and Transportation Research and Development, 2020, 14(4): 48-58.
LEI Si-da, CAO Hong-you, KANG Jun-tao. Concrete Surface Crack Recognition in Complex Scenario Based on Deep Learning. Journal of Highway and Transportation Research and Development, 2020, 14(4): 48-58.
[1] LIU Hong-gong, WANG Xue-jun, LI Bing-ying, et al. Bridge Crack Detection and Recognition based on Convolutional Neural Network[J]. Journal of Hebei University of Science and Technology, 2016, 37(5):485-490. (in Chinese) [2] CHAI Xue-song, ZHU Xing-yong, LI Jian-chao, et al. Crack Recognition Algorithm for Tunnel Lining based on Deep Convolutional Neural Network[J]. Railway Construction, 2018,58(6):60-65. (in Chinese) [3] YANG Xin-cong, LI Heng, YU Yan-tao, et al. Automatic Pixel-Level Crack Detection and Measurement Using Fully Convolutional Network[J]. Computer-Aided Civil and Infrastructure Engineering,2018,33(12):1090-1109. [4] MA Xiao-li, LU Jian. Research on Pavement Crack Image Classification Algorithm Based on Gray Analysis[J].Journal of Wuhan University of Technology:Transportation & Engineering Edition, 2018,42(5):748-752, 756. (in Chinese) [5] YUAN Wei-qi, XUE Dan. Overview of Tunnel Lining Crack Detection Algorithms Based on Machine Vision[J]. Journal of Instrumentation, 2017,38(12):3100-3111. (in Chinese) [6] LI Na, XU Yuan-fei, JIA Shu-tao. Fast Method for Bridge Crack Identification Based on Second-Order Moment and Gray Difference[J].Computer Applications and Software, 2019,36(5):216-219, 230. (in Chinese) [7] BRIGANTE M, SUMBATYAN M A. On Multiple Crack Identification by Ultrasonic Scanning[J]. Journal of Physics:Conference Series,2018,991(1):012014. [8] WANG Wei, CHEN Xin-xuan, HAN Min. Research on Improvement of Asphalt Pavement Crack Detection Technology[J].Science and Technology Information, 2019,17(8):55, 57. (in Chinese) [9] HUANG Hai-yan, LIU Xiao-ming, SUN Hua-yong, et al. Application of Cluster Analysis Algorithm in Uncertainty Decision[J].Computer Science, 2019,46(S1):593-597. (in Chinese) [10] ZHANG Chao, GUO Xiu-juan, ZHANG Kun-peng. Selection of K-means Algorithm Cluster Center[J]. Journal of Jilin University:Information Science Edition, 2019,37(4):437-441. (in Chinese) [11] YAN Sheng-long, XU Fei-hong. Crack Measurement of Concrete Structures Based on Sub-pixel Algorithm[J]. Transportation Science and Engineering, 2017,33(4):31-36. (in Chinese) [12] RUAN Xiao-li, WANG Bo, JING Guo-qiang, et al. Research on Automatic Identification of Surface Cracks in Bridge Concrete Structures[J]. World Bridges, 2017,45(6):55-59. (in Chinese) [13] LI Qing-quan, HU Qing-wu.Analysis Method of Pavement Crack Image Based on Image Automatic Homogenization[J].Highway and Transportation Science and Technology,2010,27(4):1-5, 27. (in Chinese) [14] XU Yang, BAO Yue-quan, CHEN Jia-hui. Surface Fatigue Crack Identification in Steel Box Girder of Bridges by a Deep Fusion Convolutional Neural Network Based on Consumer-grade Camera Images[J]. Stryctural Health Monitoring-an International Journal,2019(18):653-674. [15] FAN Ling.Research on Digital Image Denoising Method Based on Hybrid Filter Algorithm[J].Information Technology, 2019(08):79-82, 87. (in Chinese) [16] ALEX Rodriguez, ALESSANDRO Laio.Clustering by Fast Search and Find of Density Peaks[J].Science,2014,344:1492-1496. [17] LI Yun-hong, ZHANG Qiu-ming, ZHOU Xiao-ji, et al. Watershed Image Segmentation Algorithm Based on Morphology and Region Merger[J]. Computer Engineering and Applications, 2020, 50(2):190-195. (in Chinese) [18] GONZALEZ R C. Digital Image Processing, 3rd Edition[M]. Beijing:Electronic Industry Press, 2017. (in Chinese) [19] HAN Xiao-jian, ZHAO Zhi-cheng.Study on Detection Method of Structural Surface Cracks Based on Computer Vision Technology[J].Journal of Building Structures, 2018,39(S1):418-427. (in Chinese)
2020, Vol. 14 
