计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (11A): 240100141-9.doi: 10.11896/jsjkx.240100141

• 图像处理&多媒体技术 • 上一篇    下一篇

基于可分离卷积与小波变换融合的道路裂缝检测

刘云清1,2, 吴越1, 张琼1,2, 颜飞1,2, 陈姗姗1   

  1. 1 长春理工大学电子信息工程学院 长春 130000
    2 吉林省智能感知与信息处理科技创新中心 长春 130000
  • 出版日期:2024-11-16 发布日期:2024-11-13
  • 基金资助:
    国家自然科学基金青年科学基金(42204144);吉林省自然科学基金(YDZJ202101ZYTS064)

Road Crack Detection Based on Separable Convolution and Wave Transform Fusion

LIU Yunqing1,2, WU Yue1, ZHANG Qiong1,2, YAN Fei1,2, CHEN Shanshan1   

  1. 1 School of Electronics Information,Changchun University of Science and Technology,Changchun 130000,China
    2 Jilin Provincial Science and Technology Innovation Center of Intelligent Perception and Information Processing,Changchun 130000,China
  • Online:2024-11-16 Published:2024-11-13
  • About author:LIU Yunqing,born in 1970,professor,Ph.D supervisor.His main research interests include computer vision and radar signal processing and laser communication.
  • Supported by:
    Young Scientists Fund of the National Natural Science Foundation of China(42204144) and Natural Science Foundation of Jilin Province,China(YDZJ202101ZYTS064).

PDF (PC)

153

摘要： 针对目前对细小裂缝检测能力不强、分割精度低等问题,提出了一种改进的U-Net模型来检测路面裂缝,提高检测能力和分割精度。中文设计了新的模块MSDWBlock(Multi-Scale Depthwise Separable Convolutional Block),应用在编码器和解码器部分,通过深度可分离卷积增强模型的能力,扩大模型感受野,在跳跃连接部分引入了C2G注意力机制模块,提升模型对裂缝特征的感知能力;并引入了ASPP(Atrous Spatial Pyramid Pooling)和DWT(Discrete Wavelet Transformation)。ASPP通过在多个尺度上进行操作,有助于捕捉到裂缝的特征,而DWT能够减少卷积池化过程中的裂缝空间信息损失,保留裂缝边缘信息。这种结构设计使得网络更专注于裂缝的特征,从而提升了裂缝检测的准确性。通过实验证明所提模型显示出优于U-Net,Segnet,U2net等先进模型的精确性。在CFD数据集上mIoU,F1分别达到78.51%,0.868。这些成果表明,所提方法能有效提升道路裂缝检测的性能。

关键词: 裂缝检测, U-Net神经网络, 深度可分离卷积, 注意力机制, 空间金字塔, 小波变换

Abstract: Aiming at the current problems of weak detection ability and low segmentation accuracy for small cracks,an improved U-Net model is proposed to detect road cracks and improve detection ability and segmentation accuracy.This paper designs a new module,multi scale depth separated convolutional block(MSDWBlock),which is applied in the encoder and decoder sections.Through its depthwise separable convolution,the model's ability is enhanced,the model's receptive field is expanded,and a C2G attention mechanism module is introduced in the skip connection section to enhance the model's perception of crack features.And atrous spatial pyramid pooling(ASPP) and discrete wavelet transformation(DWT) are introduced.ASPP helps to capture the characteristics of cracks by operating at multiple scales,while DWT can reduce the loss of crack spatial information during convolutional pooling and preserve crack edge information.This structural design makes the network more focused on the characteristics of cracks,thereby improving the accuracy of crack detection.It has been demonstrated through experiments that the accuracy of the proposed model is better than that of advanced models such as U-Net,Segnet,and U2net.On the CFD dataset,mIoU and F1 reaches 78.51% and 0.868 respectively.These results indicate that the proposed method can effectively improve the perfor-mance of road crack detection.

Key words: Crack detection, U-net neural network, Depthwise separable convolutional, Attention mechanism, Spatial pyramid, Wavelet transform

中图分类号: 

  • TP391
[1]2020 Statistical Bulletin on the Development of the Transportation Industry [J].Communications Finance and Accounting,2021(6):92-97.
[2]MA L,LI J.SD-GCN:Saliency-based dilated graph convolutionnetwork for pavement crack extraction from 3D point clouds[J].International Journal of Applied Earth Observation and Geoinformation,2022,111:102836.
[3]GUPTA P,DIXIT M.Image-based crack detection approaches:a comprehensive survey[J].Multimedia Tools and Applications,2022,81(28):40181-40229.
[4]SHI Y,CUI L,QI Z,et al.Automatic road crack detec-tion using random structured forests[J].IEEE Transactions on Intelligent Transportation Systems,2016,17(12):3434-3445.
[5]ZOU Q,CAO Y,LI Q,et al.CrackTree:Automatic crack detection from pavement images[J].Pattern Recognition Letters,2012,33(3):227-238.
[6]LI Q,ZOU Q,ZHANG D,et al.FoSA:F*seed-growing ap-proach for crack-line detection from pavement images[J].Image and Vision Computing,2011,29(12):861-872.
[7]DORAFSHAN S,THOMAS R J,MAGUIREM.Comparison ofdeep convolutional neural networks and edge detectors for image-based crack detection in concrete[J].Construction and Building Materials,2018,186:1031-1045.
[8]KAMALIARDAKANI M,SUN L,ARDAKANI M K.Sealed-crack detection algorithm using heuristic thresholding approach[J].Journal of Computing in Civil Engineering,2016,30(1):04014110.
[9]WANG J,LIU F,YANG W,et al.Pavement crack detectionusing attention u-net with multiple sources[C]//Pattern Recognition and Computer Vision:Third Chinese Conference(PRCV 2020).Nanjing,China,Part II 3.Springer International Publishing,2020:664-672.
[10]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).IEEE,2016:3708-3712.
[11]ZHANG A,WANG K C P,LI B,et al.Automated pixel-levelpavement crack detection on 3D asphalt surfaces using a deep-learning network[J].Computer-Aided Civil and Infrastructure Engineering,2017,32(10):805-819.
[12]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.
[13]ZHU S,DU J,LI Y,et al.Method for bridge crack detectionbased on the U-Net convolutional networks[J].Journal of Xi-dian University,2019,46(4):35-42.
[14]LIU Z,CAO Y,WANG Y,et al.Computer vision-based concretecrack detection using U-net fully convolutional networks[J].Automation in Construction,2019,104:129-139.
[15]HUYAN J,LI W,TIGHE S,et al.CrackU-net:A novel deepconvolutional neural network for pixelwise pavement crack detection[J].Structural Control and Health Monitoring,2020,27(8):e2551.
[16]HOWARD A G,ZHU M,CHEN B,et al.Mobilenets:Efficient convolutional neural networks for mobile vision applications[J].arXiv:1704.04861,2017.
[17]GUO T,MOUSAVI H S,VU T H,et al.Deep wavelet prediction for image super-resolution[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2017:104-113.
[18]BAE W,YOO J,CHULYE J.Beyond deep residual learning for image restoration:Persistent homology-guided manifold simplification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops.2017:145-153.
[19]LI Q,SHEN L,GUO S,et al.Wavelet integrated CNNs fornoise-robust image classification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2020:7245-7254.
[20]MNIH V,BADIA A P,MIRZA M,et al.Asynchronous methods for deep reinforcement learning[C]//International Conference on Machine Learning.PMLR,2016:1928-1937.
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