计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (6A): 220300214-6.doi: 10.11896/jsjkx.220300214

• 图像处理&多媒体技术 • 上一篇    下一篇

基于注意力机制和可变形卷积的路面裂缝检测

隆涛1, 董安国1, 刘来君2   

  1. 1 长安大学理学院 西安 710064;
    2 长安大学公路学院 西安 710064
  • 出版日期:2023-06-10 发布日期:2023-06-12
  • 通讯作者: 隆涛(longtao1220@163.com)
  • 基金资助:
    陕西省重点产业创新链项目(2020ZDLGY09-09);国家自然科学基金青年项目(12001057)

Pavement Crack Detection Based on Attention Mechanism and Deformable Convolution

LONG Tao1, DONG Anguo1, LIU Laijun2   

  1. 1 School of Science,Chang’an University,Xi’an 710064,China;
    2 School of Highway,Chang’an University,Xi’an 710064,China
  • Online:2023-06-10 Published:2023-06-12
  • About author:LONG Tao,born in 1997,postgraduate.His main research interests include computer vision and crack detection based on deep learning.
  • Supported by:
    Key Industry Innovation Chain Project of Shaanxi,China(2020ZDLGY09-09) and Youth Project of National Na-tural Science Foundation of China(12001057).

PDF (PC)

298

摘要： 针对较复杂背景下路面裂缝检测问题,由于基于深度学习的图像分割算法检测效果不甚理想,以及裂缝图像自身像素类别不平衡,提出了一种基于注意力机制和可变形卷积的路面裂缝检测网络,该网络基于编码-解码结构进行构建。为了解决较为复杂背景裂缝检测困难的问题,首先,由可变形卷积提升网络对不同形状裂缝线性特征的学习能力;其次,使用密集连接机制强化特征信息;然后,在解码阶段采用转置卷积和桥接方式与编码阶段特征逐步融合,并结合多级特征融合的思想,提高网络的检测精度;最后,引入注意力模块(SimAM),在不增加网络参数的前提下,更加关注目标特征的提取,抑制背景特征。在两个公开裂缝数据集上进行实验来验证该算法的有效性,实验结果表明,该算法的各项性能评价指标均优于对比算法,BCrack数据集的平均像素精度、平均交并比分别达到92.12%和84.79%,CFD数据集的平均像素精度、平均交并比分别达到91.02%和74.75%,在复杂背景裂缝检测下表现良好,可应用于路面维修工程。

关键词: 裂缝检测, 编码-解码结构, 可变形卷积, 密集连接机制, 注意力模块

Abstract: Aiming at the pavement crack detection problem under complex background,due to the unsatisfactory detection effect of image segmentation algorithm based on deep learning,and the imbalance of pixel categories in the crack image itself,this paper proposes a pavement crack detection network based on attention mechanism and deformable convolution,which is constructed based on encoder-decoder structure.In order to solve the problem of difficult crack detection in complex background,firstly,deformable convolutional is used to improve the learning ability of linear features of cracks with different shapes.Secondly,the dense connection mechanism is used to strengthen the feature information.Then,in the decoder stage,the feature fusion of transpose convolution and bridge are adopted,and the multi-stage feature fusion is combined to improve the detection accuracy of the network.Finally,the attention module(SimAM) is introduced to pay more attention to the extraction of target features and suppress background features without increasing network parameters.Experiments are carried out on two open crack datasets to ve-rify the effectiveness of the algorithm.The experimental results show that the performance evaluation criteria of the algorithm are better than the comparison algorithms.The mean pixel accuracy and mean intersection over union of the BCrack dataset reached 92.12% and 84.79%,respectively.The mean pixel accuracy and mean intersection over union of the CFD dataset reached 91.02% and 74.75%,respectively.The average accuracy and average intersection ratio of CFD data set is 91.02% and 74.75%,respectively.The algorithm performs well in crack detection under complex background,and can be applied to pavement maintenance engineering.

Key words: Crack detection, Encoder-decoder structure, Deformable convolutional, Dense connection mechanism, Attention module

中图分类号: 

  • TP391
[1]XU H,SU X,WANG Y,et al.Automatic bridge crack detection using a convolutional neural network[J].Applied Sciences,2019,9(14):2867.
[2]CHENG Y,TIAN L,YIN C,et al.A magnetic domain spots fil-tering method with self-adapting threshold value selecting for crack detection based on the MOI[J].Nonlinear Dynamics,2016,86(2):741-750.
[3]XU L L,ZHU X P,HOU Y X,et al.Culvert crack defect segmentation algorithm based on enhanced tone features[J].Progress in Laser and Optoelectronics,2020,57(8):081016.
[4]HOANG N D,NGUYEN Q L,TRAN V D.Automatic recognition of asphalt pavement cracks usingmetaheuristic optimized edge detection algorithms and convolution neural network[J].Automation in Construction,2018,94:203-213.
[5]CHA Y J,CHOI W,BUYUKOZTURK O.Deep learning-based crack damage detection using convolutional neural networks[J].Computer-Aided Civil and Infrastructure Engineering,2017,32(5):361-378.
[6]LI F L,Ma W F,LI L,et al.Research on Bridge Crack Detection Algorithm Based on Deep Learning[J].Acta Automatica Sinica,2019,45(9):1727-1742.
[7]ZHANG Y,HUANG J,CAI F.On Bridge Surface Crack Detec-tion Based on an Improved YOLO v3Algorithm[J].IFAC-PapersOnLine,2020,53(2):8205-8210.
[8]LECUN Y,BENGIO Y,HINTON G.Deep learning[J].nature,2015,521(7553):436-444.
[9]BADRINARAYANAN V,KENDALL A,CIPOLLA R.Seg-Net:A deep convolutional encoder-decoder architecture for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(12):2481-2495.
[10]ZHAO H,SHI J,QI X,et al.Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:6230-6239.
[11]CHEN L C,PAPANDREOU G,KOKKINOS I,et al.Deeplab:Semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected crfs[J].IEEE transactions on pattern analysis and machine intelligence,2017,40(4):834-848.
[12]RONNEBERGER O,FISCHER P,BROX T.U-net:Convolu-tional networks for biomedical image segmentation[C]//International Conference on Medical Image Computing and Compu-ter-assisted Intervention.2015:234-241.
[13]YANG X,LI H,YU Y,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.
[14]REN Y,HUANG J,HONG Z,et al.Image-based concrete crack detection in tunnels using deep fully convolutional networks[J].Construction and Building Materials,2020,234:117367.
[15]ZHOU Q,QU Z,CAO C.Mixed pooling and richer attention feature fusion for crack detection[J].Pattern Recognition Letters,2021,145:96-102.
[16]QIAO W,LIU Q,WU X,et al.Automatic pixel-level pavement crack recognition using a deep feature aggregation segmentation network with a scSE attention mechanism module[J].Sensors,2021,21(9):2902.
[17]LIU J,YANG X,LAU S,et al.Automated pavement crack detection and segmentation based on two-step convolutional neural network[J].Computer-Aided Civil and Infrastructure Enginee-ring,2020,35(11):1291-1305.
[18]ZHANG L X,SHEN J K,ZHU B J.A research on an improved Unet-based concrete crack detection algorithm[J].Structural Health Monitoring,2021,20(4):1864-1879.
[19]GAO H,YUAN H,WANG Z,et al.Pixel transposed convolutional networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2019,42(5):1218-1227.
[20]CAI S,SHU Y,CHEN G,et al.Effective and efficient dropout for deep convolutional neural networks[J].arXiv,2019,1904.03392.
[21]LIU Y,WANG W,Li Q,et al.DCNet:A deformable convolutional cloud detection network for remote sensing imagery[J].IEEE Geoscience and Remote Sensing Letters,2021,19(1):1-5.
[22]YANG L,ZHANG R Y,LI L,et al.Simam:A simple,parameter-free attention module for convolutional neural networks[C]//Proceedings of Machine Learning Research.2021:11863-11874.
[23]LI G,MA B,HE S,et al.Automatic tunnel crack detection based on u-net and a convolutional neural network with alternately updated clique[J].Sensors,2020,20(3):717.
[24]EELBODE T,BERTELS J,BERMAN M,et al.Optimization for medical image segmentation:theory and practice when evaluating with Dice score or Jaccard index[J].IEEE Transactions on Medical Imaging,2020,39(11):3679-3690.
[25]LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2020:318-327.
[26]FU H X,MENG D,LI W H,et al.Bridge Crack Semantic Segmentation Based on Improved Deeplabv3+[J].Journal of Marine Science and Engineering,2021,9(6):671.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发