| 322 | 2 | 84 |
| 下载次数 | 被引频次 | 阅读次数 |
针对现有方法存在的隧道渗水检测精度不高和特征融合过程中信息丢失的问题，提出了一种基于有效的通道注意力（efficient channel attention,ECA）和YOLOv5的隧道渗水检测方法。该方法融合ECA注意力模块设计瓶颈结构，加强挖掘浅层特征表征的几何结构信息，充分提取水迹特征信息的同时抑制背景特征，提高水迹检测精度。在建立的隧道渗水水迹数据集上进行实验，结果表明：对比原YOLOv5模型，所提出的隧道渗水检测方法的平均精度均值提高了10%，准确率提高了17%，召回率提高了6%。实验结果验证了该方法的有效性。
Abstract：To solve the problems of low detection accuracy and information loss in feature fusion process for methods of the existing tunnel water seepage detection, a tunnel water seepage detection method based on ECA( efficient channel at ten tion, ECA) and YOLOv5 is proposed, which fuses the ECA attention module to design the bottleneck structure, enhances the mining of geometric structure information with shallow feature representation, fully extracts water stain feature information while suppressing background features, and improves water stain detection accuracy. Experimental results using an established water trace data set of tunnel seepage demonstrate that the proposed method outperforms the original YOLOv5model. Specifically, the proposed tunnel seepage detection method achieves a 10% increase in average accuracy, a 17% increase in accuracy rate, and a 6% increase in recall rate, validating its effectiveness in tunnel seepage detection.
[1] XUE Y D,CAI X Y,SHADABFAR M,et al. Deep learningbased automatic recognition of water leakage area in shield tunnel lining[J]. Tunnelling and Underground Space Technology,2020,104:103524.
[2]龚起民.车载式隧道表面图像采集与病害检测技术[D].北京：北京交通大学，2020.
[3] KIM H,AHN E,CHO S,et al. Comparative analysis of image binarization methods for crack identification in concrete structures[J]. Cement and Concrete Research,2017,99:53-61.
[4]杨骏浩,傅丹妮,陈盈娜.数字图像处理技术在织物瑕疵检测中的应用研究[J].丝网印刷, 2023(15):23-25.
[5] YU Z W,SHEN Y G,SHEN C K. A real-time detection approach for bridge cracks based on YOLOv4-FPM[J]. Automation in Construction,2021,122:103514.
[6] LIU W,ANGUELOV D,ERHAN D,et al. SSD:single shot multiBox detector[C]//European Conference on Computer Vision. Cham:Springer,2016:21-37.
[7] REDMON J,FARHADI A. YOLO9000:bettter,faster,stronger[C]//2017 IEE E Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE,2017:6517-6525.
[8] 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). USA:IEEE,2016:779-788.
[9] ZHANG S S, YANG J A, SCHIELE B. Occluded pedestrian detection through guided attention in CNNs[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City:IEEE, 2018:6995-7003.
[10]蒋博,万毅,谢显中.改进YOLOv5s的轻量化钢材表面缺陷检测模型[J].计算机科学, 2023, 50(S2):271-277.
[11]郭磊,王洋,靳正轩，等.基于改进Faster RCNN的化纤丝缺陷检测[J].毛纺科技, 2023, 51(3):74-79.
[12]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.
[13]ZHAO L,WANG J W,LIU S P,et al. An adaptive multitask network for detecting the region of water leakage in tunnels[J].Applied Sciences,2023,13(10):6231.
[14] LI J Q,HE Z Y,LI D X,et al. Research on water seepage detection technology of tunnel asphalt pavement based on deep learning and digital image processing[J]. Scientific Reports,2022,12(1):11519.
[15]田有良，樊廷立，唐超.面向地铁隧道表面渗漏水的快速检测技术[J].测绘通报，2022(9):29-33.
[16]周宝定，谢沛瑶，郭文浩，等.基于激光点云灰度图像的隧道渗水病害检测[J].测绘通报，2023(8):34-39,90.
[17]LI G P,CHEN S D,WANG L,et al. Vehicle-side target detection based on improved YOLOv5[J]. Computer Systems and Applications,2022,31(12):127-134.
[18]WANG Q,WU B,ZHU P,et al. ECA-net:efficient channel attention for deep convolutional neural networks[C]//2020IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). USA:IEEE,2020:11534-11542.
基本信息:
DOI：10.19573/j.issn2095-0926.202402003
中图分类号:TP391.41;TP18;U457.2
引用信息:
[1]杨丽,邓靖威,段海龙,等.基于ECA和YOLOv5的隧道渗水检测方法[J].天津职业技术师范大学学报,2024,34(02):19-24.DOI:10.19573/j.issn2095-0926.202402003.
基金信息:
天津市教委科研计划项目（2022ZD036）; 天津市自然科学基金资助项目（20JCZDJC00150）
2024-01-14
2024
2024-04-25
2024-06-14
2024
1
2024-06-26
2024-06-26