•  
  •  
 

Abstract

In view of the existing problems in pavement crack segmentation, this paper proposed a Crack Mask R‑CNN pixel-level segmentation algorithm. Crack Mask R‑CNN is a case segmentation framework for pavement crack images, which can not only detect cracks in the image but also give a high-quality segmentation result for the specific contour of each crack. Firstly, the big data of collected road cracks was denoised and enhanced, and the data set for model training and testing was constructed. Secondly, by optimizing the proportion and size of anchor frames in the segmentation algorithm, the accuracy of the model in selecting crack candidate regions was improved, and the IoU‑guided non‑maximum suppression (NMS) algorithm was used to replace the traditional algorithm, so as to improve the segmentation accuracy of road cracks. In the aspect of model learning-based hyperparameter optimization, by training a variety of combination examples, the hyperparameter combination with the best segmentation effect was selected, and finally, the segmentation model with a segmentation accuracy of 93.45% was trained. Finally, by extracting the topological feature information of the crack region, the pixel-size information of the crack could be effectively measured.

Publication Date

11-24-2023

DOI

10.14048/j.issn.1671-2579.2023.05.009

First Page

47

Last Page

52

Submission Date

March 2025

Reference

[1] CHEN J, WU J J, CHEN G, et al. Design and development of a multi-rotor unmanned aerial vehicle system for bridge inspection[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2016: 498-510. [2] 周基, 蔡强, 田琼. 70年中国公路路基路面病害研究现状与发展趋势: 基于CNKI1949—2019年文献的知识图谱分析[J]. 中外公路, 2020, 40(3): 60-66. ZHOU Ji, CAI Qiang, TIAN Qiong. Research status and development trend of highway subgrade and pavement diseases in China in 70 years : knowledge map analysis based on CNKI 1949-2019 literature[J]. Journal of China & Foreign Highway, 2020, 40(3): 60-66. [3] CAO W M, LIU Q F, HE Z Q. Review of pavement defect detection methods[J]. IEEE Access, 2020, 8: 14531-14544. [4] ZOU Q, CAO Y, LI Q Q, et al. CrackTree: Automatic crack detection from pavement images[J]. Pattern Recognition Letters, 2012, 33(3): 227-238. [5] WANG S C, TANG W S. Pavement crack segmentation algorithm based on local optimal threshold of cracks density distribution[M]//Lecture Notes in Computer Science. Berlin, Heidelberg: Springer Berlin Heidelberg, 2011: 298-302. [6] MUDULI P R, PATI U C. A novel technique for wall crack detection using image fusion[C]//2013 International Conference on Computer Communication and Informatics. Coimbatore, India. IEEE, 2013: 1-6. [7] AYENU-PRAH A, ATTOH-OKINE N. Evaluating pavement cracks with bidimensional empirical mode decomposition[J]. EURASIP Journal on Advances in Signal Processing, 2008, 2008(1): 861701. [8] CORREIA P, ROSA P. Automatic road pavement crack detection using boosting classifiers[Z], 2009. [9] AKAGIC A, BUZA E, OMANOVIC S, et al. Pavement crack detection using Otsu thresholding for image segmentation[C]//2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO). Opatija, Croatia. IEEE, 2018: 1092-1097. [10] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, NV, USA. IEEE, 2016: 779-788. [11] LIU W, ANGUELOV D, ERHAN D, et al. SSD: single shot MultiBox detector[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2016: 21-37. [12] ZHANG L, YANG F, ZHANG Y D, et al. Road crack detection using deep convolutional neural network[C]//2016 IEEE International Conference on Image Processing (ICIP). Phoenix, AZ, USA. IEEE, 2016: 3708-3712. [13] CHA Y J, CHOI W, BÜYÜKÖZTÜRK O. Deep learning-based crack damage detection using convolutional neural networks[J]. Computer-Aided Civil and Infrastructure Engineering, 2017, 32(5): 361-378. [14] ZOU Q, ZHANG Z, LI Q Q, et al. DeepCrack: learning hierarchical convolutional features for crack detection[J]. IEEE Transactions on Image Processing, 2018: 361-378. [15] DAVID JENKINS M, CARR T A, IGLESIAS M I, et al. A deep convolutional neural network for semantic pixel-wise segmentation of road and pavement surface cracks[C]//2018 26th European Signal Processing Conference (EUSIPCO). Rome, Italy. IEEE, 2018: 2120-2124. [16] YANG X C, LI H, YU Y T, 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. [17] JIANG B R, LUO R X, MAO J Y, et al. Acquisition of localization confidence for accurate object detection[M]//Lecture Notes in Computer Science. Cham: Springer International Publishing, 2018: 816-832. [18] LIN T Y, DOLLÁR P, GIRSHICK R, et al. Feature pyramid networks for object detection[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA. IEEE, 2017: 936-944. [19] REN S Q, HE K M, GIRSHICK R, et al. Faster R-CNN: towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149. [20] 翁飘, 陆彦辉, 齐宪标, 等. 基于改进的全卷积神经网络的路面裂缝分割技术[J]. 计算机工程与应用, 2019, 55(16): 235-239+245. WENG Piao, LU Yanhui, QI Xianbiao, et al. Pavement crack segmentation technology based on improved fully convolutional networks[J]. Computer Engineering and Applications, 2019, 55(16): 235-239+245.

Download

Share

COinS