Deep learning-based algorithm for multi defect detection in tunnel lining

doi:10.3785/j.issn.1008-973X.2024.06.007

Journal of ZheJiang University (Engineering Science)

2024, Vol. 58

Issue (6): 1161-1173 DOI: 10.3785/j.issn.1008-973X.2024.06.007

Deep learning-based algorithm for multi defect detection in tunnel lining

Juan SONG(

),Longxi HE*(

),Huiping LONG

School of Civil and Architectural Engineering, Shaoyang College, Shaoyang 422000, China

Download:

HTML

PDF(4841KB) HTML
Export: BibTeX | EndNote (RIS)

Abstract

A tunnel lining surface defect detection algorithm TDD-YOLO was proposed, for the problems of insufficient global information extraction and low detection accuracy of existing object detection algorithms in tunnel lining defect detection. The algorithm was based on the YOLOv7 framework. Firstly, MobileViT was used as the backbone feature extraction network to improve the global and local information extraction capability of the network. Secondly, Coordinate attention (CA) module was added after the upsampling and downsampling of the feature pyramid network to highlight the feature information of defects and remove the interference of background information. Finally, a convolutional module called TP Block was proposed to further improve the feature extraction ability of the network with less computation. Five algorithms, SSD, Faster-RCNN, EfficientDet, YOLOv5 and YOLOv7, were selected for comparison and analysis, in order to verify the effectiveness of the proposed algorithm. Results showed that the F1 of TDD-YOLO algorithm was 77.43%, which had an improvement of 15.58%, 17.36%, 12.19%, 6.32%, and 6.14%, respectively, compared with the above five contrast algorithms. The mAP was 77.52%, which had an improvement of 15.20%, 14.24%, 9.44%, 7.44%, and 6.39%, respectively. The TDD-YOLO algorithm has the highest defect recognition accuracy and the best overall performance, which is suitable for defect detection task of actual tunnel projects.

Key words： tunnel defect deep learning defect recognition object detection neural network

Received: 04 June 2023 Published: 25 May 2024

CLC:

TP 393

Fund: 湖南省教育厅一般项目（20C1658，21C0602）.

Corresponding Authors: Longxi HE E-mail: songjsyuniversity@163.com;helxshaoyang@163.com

	Service
	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Juan SONG
	Longxi HE
	Huiping LONG

Cite this article:

Juan SONG,Longxi HE,Huiping LONG. Deep learning-based algorithm for multi defect detection in tunnel lining. Journal of ZheJiang University (Engineering Science), 2024, 58(6): 1161-1173.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2024.06.007 OR https://www.zjujournals.com/eng/Y2024/V58/I6/1161

基于深度学习的隧道衬砌多病害检测算法

针对已有目标检测算法在隧道衬砌病害检测中全局信息提取不充分、检测精度低的问题，提出隧道衬砌表观病害检测算法TDD-YOLO. 该算法以YOLOv7框架为基础，采用MobileViT作为主干特征提取网络，提高网络全局信息和局部信息提取能力；在特征金字塔网络的上采样和下采样后增加Coordinate attention (CA)注意力模块，突出病害的特征信息，去除背景信息的干扰；提出卷积模块TP Block，在计算量较小的情况下进一步提高网络的特征提取能力. 为了验证所提出算法的有效性，选用SSD、Faster-RCNN、EfficientDet、YOLOv5、YOLOv7这5种算法进行对比分析. 实验结果表明，TDD-YOLO算法的F1为77.43%，相对5种对比算法，分别提高了15.58%、17.36%、12.19%、6.32%、6.14%；mAP为77.52%，相对5种对比算法，分别提高了15.20%、14.24%、9.44%、7.44%、6.39%. TDD-YOLO算法病害识别精度最高，综合性能最优，适用于实际隧道工程的病害检测任务.

关键词： 隧道病害, 深度学习, 病害识别, 目标检测, 神经网络

Fig.1 YOLOv7 network structure

Fig.2 Multi-branch stacked module structure

Fig.3 Transition module structure

Fig.4 MobileViT network architecture

Fig.5 Coordinate attention module structure

Fig.6 Conventional convolution and partial convolution structures

Fig.7 TDD-YOLO network structure

Fig.8 Schematic diagram of different sizes of priori frames

Fig.9 Example of tunnel apparent disease image

Fig.10 Disease image enhancement method

Tab.1 Parameters of tunnel defect data set

Fig.11 Disease labeling image

Tab.2 Network training hyperparameters

Tab.3 Backbone network evaluation indicators

Tab.4 Network ablation experiment results

Tab.5 Accuracy experimental result comparison of six models

Fig.12 Tunnel defect image detection results in simple and bright environment

Fig.13 Tunnel defect image detection results in complex and somber environment


[1]	《中国公路学报》编辑部中国交通隧道工程学术研究综述: 2022[J]. 中国公路学报, 2022, 35 (4): 1- 40 “China Journal of Highway and Transport” editorial department Review on china's traffic tunnel engineering research: 2022[J]. China Journal of Highway and Transport, 2022, 35 (4): 1- 40 doi: 10.3969/j.issn.1001-7372.2022.04.001

[2]	董飞, 房倩, 张顶立, 等北京地铁运营隧道病害状态分析[J]. 土木工程学报, 2017, 50 (6): 104- 113 DONG Fei, FANG Qian, ZHANG Dingli, et al Analysis on defects of operational metro tunnels in Beijing[J]. China Civil Engineering Journal, 2017, 50 (6): 104- 113

[3]	HUANG H, LI Q, ZHANG D Deep learning based image recognition for crack and leakage defects of metro shield tunnel[J]. Tunnelling and Underground Space Technology, 2018, 77: 166- 176 doi: 10.1016/j.tust.2018.04.002

[4]	许力之, 王耀东, 朱力强, 等隧道表面图像多目标智能识别算法研究[J]. 铁道学报, 2022, 44 (9): 154- 162 XU Lizhi, WANG Yaodong, ZHU Liqiang, et al Research on algorithm of intelligent recognition of mutiple objects of tunnel surface image[J]. Journal of the China Railway Society, 2022, 44 (9): 154- 162 doi: 10.3969/j.issn.1001-8360.2022.09.020

[5]	周鸣亮, 程文, 张东明, 等运营期盾构隧道结构病害的自动化检测与三维可视化[J]. 应用基础与工程科学学报, 2021, 29 (5): 1265- 1279 ZHOU Mingliang, CHENG Wen, ZHANG Dongming, et al Automatic detection and 3D visualization of structural diseases of shield tunnel during operation[J]. Journal of Basic Science and Engineering, 2021, 29 (5): 1265- 1279

[6]	MA D, LIU J, FANG H, et al A multi-defect detection system for sewer pipelines based on StyleGAN-SDM and fusion CNN[J]. Construction and Building Materials, 2021, 312: 125385 doi: 10.1016/j.conbuildmat.2021.125385

[7]	ROSTEN E, PORTER R, DRUMMOND T Faster and better: a machine learning approach to corner detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 32 (1): 105- 119

[8]	DONG Y, WANG J, WANG Z, et al A deep-learning-based multiple defect detection method for tunnel lining damages[J]. IEEE Access, 2019, 7: 182643- 182657 doi: 10.1109/ACCESS.2019.2931074

[9]	雷明锋, 张运波, 王卫东, 等岩石岩性Mask R-CNN智能识别方法与应用研究[J]. 铁道科学与工程学报, 2022, 19 (11): 3372- 3382 LEI Mingfeng, ZHANG Yunbo, WANG Weidong, et al Investigation and application on lithology intelligent recognition method based on Mask R-CNN[J]. Journal of Railway Science and Engineering, 2022, 19 (11): 3372- 3382

[10]	XUE Y, CAI X, SHADABFAR M, et al Deep learning-based automatic recognition of water leakage area in shield tunnel lining[J]. Tunnelling and Underground Space Technology, 2020, 104: 103524 doi: 10.1016/j.tust.2020.103524

[11]	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 doi: 10.1016/j.conbuildmat.2019.117367

[12]	SAVINO P, TONDOLO F Automated classification of civil structure defects based on convolutional neural network[J]. Frontiers of Structural and Civil Engineering, 2021, 15 (2): 305- 317 doi: 10.1007/s11709-021-0725-9

[13]	DUNG C V Autonomous concrete crack detection using deep fully convolutional neural network[J]. Automation in Construction, 2019, 99: 52- 58 doi: 10.1016/j.autcon.2018.11.028

[14]	MANDAL V, UONG L, ADU-GYAMFI Y. Automated road crack detection using deep convolutional neural networks [C]// IEEE International Conference on Big Data . IEEE, 2018: 5212-5215.

[15]	CHENG J C P, WANG M Automated detection of sewer pipe defects in closed-circuit television images using deep learning techniques[J]. Automation in Construction, 2018, 95: 155- 171 doi: 10.1016/j.autcon.2018.08.006

[16]	周中, 张俊杰, 龚琛杰, 等基于深度语义分割的隧道渗漏水智能识别[J]. 岩石力学与工程学报, 2022, 41 (10): 2082- 2093 ZHOU Zhong, ZHANG Junjie, GONG Chenjie, et al Automatic identification of tunnel leakage based on deep semantic segmentation[J]. Chinese Journal of Rock Mechanics and Engineering, 2022, 41 (10): 2082- 2093

[17]	彭磊, 周春, 胡锋, 等基于深度学习的隧道掌子面节理智能检测与分割[J]. 人民长江, 2023, 54 (10): 243- 250 PENG Lei, ZHOU Chun, HU Feng, et al Research on intelligent detection and segmentation of rock joints tunnel face based on deep learning[J]. Yangtze River, 2023, 54 (10): 243- 250

[18]	刘德军, 仲飞, 黄宏伟, 等运营隧道衬砌病害诊治的现状与发展[J]. 中国公路学报, 2021, 34 (11): 178- 199 LIU Dejun, ZHONG Fei, HUANG Hongwei, et al Present Status and development trend of diagnosis and treatment of tunnel lining diseases[J]. China Journal of Highway and Transport, 2021, 34 (11): 178- 199 doi: 10.3969/j.issn.1001-7372.2021.11.015

[19]	路耀邦, 刘永胜, 樊晓东地铁隧道结构表观病害快速检测方法与应用[J]. 隧道建设: 中英文, 2021, 41 (Suppl.2): 655- 663 LU Yaobang, LIU Yongsheng, FAN Xiaodong Rapid detection method for surface defect for metro tunnel structure and its Application[J]. Tunnel Construction, 2021, 41 (Suppl.2): 655- 663

[20]	薛亚东, 李宜城基于深度学习的盾构隧道衬砌病害识别方法[J]. 湖南大学学报: 自然科学版, 2018, 45 (3): 100- 109 XUE Yadong, LI Yicheng A method of disease recognition for shield tunnel lining based on deep learning[J]. Journal of Hunan University: Natural Sciences, 2018, 45 (3): 100- 109

[21]	LIU J, ZHAO Z, LV C, et al An image enhancement algorithm to improve road tunnel crack transfer detection[J]. Construction and Building Materials, 2022, 348: 128583 doi: 10.1016/j.conbuildmat.2022.128583

[22]	ZHOU Z, ZHANG J, GONG C Automatic detection method of tunnel lining multi-defects via an enhanced You Only Look Once network[J]. Computer-Aided Civil and Infrastructure Engineering, 2022, 37 (6): 762- 780 doi: 10.1111/mice.12836

[23]	LI Y, BAO T, LI T, et al A robust real-time method for identifying hydraulic tunnel structural defects using deep learning and computer vision[J]. Computer-Aided Civil and Infrastructure Engineering, 2023, 38 (10): 1381- 1399

[24]	王宝坤, 王如路, 陈锦剑, 等. 基于深度学习的盾构隧道表观病害自动检测方法[EB/OL]. (2023-06-13) [2024-04-21]. https://doi.org/10.16183/j.cnki.jsjtu.2023.089.

[25]	XU Y, LI D, XIE Q, et al Automatic defect detection and segmentation of tunnel surface using modified Mask R-CNN[J]. Measurement, 2021, 178: 109316 doi: 10.1016/j.measurement.2021.109316

[26]	WANG C Y, BOCHKOVSKIY A, LIAO H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway: IEEE, 2023: 7464−7475.

[27]	MAKANTASIS K, PROTOPAPADAKIS E, DOULAMIS A, et al. Deep convolutional neural networks for efficient vision based tunnel inspection [C]// 2015 IEEE International Conference on Intelligent Computer Communication and Processing . [s. l.]: IEEE, 2015: 335−342.

[28]	MEHTA S, RASTEGARI M. Mobilevit: light-weight, general-purpose, and mobile-friendly vision transformer [EB/OL]. [2024-04-21]. https://arxiv.longhoe.net/abs/2110.02178.

[29]	HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway: IEEE: 2021: 13713−13722.

[30]	CHEN J, KAO S, HE H, et al. Run, don't walk: chasing higher FLOPS for faster neural networks [C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . Piscataway: IEEE, 2023: 12021−12031.

[31]	MENG Z, XU S, WANG L, et al Defect object detection algorithm for electroluminescence image defects of photovoltaic modules based on deep learning[J]. Energy Science and Engineering, 2022, 10 (3): 800- 813 doi: 10.1002/ese3.1056

[32]	WANG Z, CAI Z, WU Y An improved YOLOX approach for low-light and small object detection: PPE on tunnel construction sites[J]. Journal of Computational Design and Engineering, 2023, 10 (3): 1158- 1175 doi: 10.1093/jcde/qwad042

[33]	李明超, 符家科, 张野, 等耦合岩石图像与锤击音频的岩性分类智能识别分析方法[J]. 岩石力学与工程学报, 2020, 39 (5): 996- 1004 LI Mingchao, FU Jiake, ZHANG Ye, et al Intelligent recognition and analysis method of rock lithology classification based on coupled rock images and hammering audios[J]. Chinese Journal of Rock Mechanics and Engineering, 2020, 39 (5): 996- 1004

[1]	Yidan LIU,Xiaofei ZHU,Yabo YIN. Heterogeneous graph convolutional neural network for argument pair extraction[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(5): 900-907.

[2]	Zhiwei XING,Shujie ZHU,Biao LI. Airline baggage feature perception based on improved graph convolutional neural network[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(5): 941-950.

[3]	Cuiting WEI,Weijian ZHAO,Bochao SUN,Yunyi LIU. Intelligent rebar inspection based on improved Mask R-CNN and stereo vision[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(5): 1009-1019.

[4]	Bo ZHONG,Pengfei WANG,Yiqiao WANG,Xiaoling WANG. Survey of deep learning based EEG data analysis technology[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(5): 879-890.

[5]	Hai HUAN,Yu SHENG,Chenxi GU. Global guidance multi-feature fusion network based on remote sensing image road extraction[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(4): 696-707.

[6]	Xianglong LUO,Yafei WANG,Yanbo WANG,Lixin WANG. Structural deformation prediction of monitoring data based on bi-directional gate board learning system[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(4): 729-736.

[7]	Tianmin DENG,Xinxin CHENG,Jinfeng LIU,Xiyue ZHANG. Small target detection algorithm for aerial images based on feature reuse mechanism[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 437-448.

[8]	Zhuo WANG,Yongqiang LI,Yu FENG,Yuanjing FENG. Policy gradient algorithm and its convergence analysis for two-player zero-sum Markov games[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 480-491.

[9]	Yu ZHOU,Luyi GAN,Shengkui DI,Wenyu HE,Ningbo LI. Bridge model modification experiment based on strain influence line[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 537-546.

[10]	Mingjun SONG,Wen YAN,Yizhao DENG,Junran ZHANG,Haiyan TU. Light-weight algorithm for real-time robotic grasp detection[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 599-610.

[11]	Juncheng WANG,Fahui WANG. Variable weight PSO-Elman neural network based roadadhesion coefficient estimation[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 622-634.

[12]	Qingjie QIAN,Junhe YU,Hongfei ZHAN,Rui WANG,Jian HU. Dimension prediction method of injection molded parts based on multi-feature fusion of DL-BiGRU[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(3): 646-654.

[13]	Xinhua YAO,Tao YU,Senwen FENG,Zijian MA,Congcong LUAN,Hongyao SHEN. Recognition method of parts machining features based on graph neural network[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(2): 349-359.

[14]	Siyi QIN,Shaoyan GAI,Feipeng DA. Video object detection algorithm based on multi-level feature aggregation under mixed sampler[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(1): 10-19.

[15]	Xuefei SUN,Ruifeng ZHANG,Xin GUAN,Qiang LI. Lightweight and efficient human pose estimation with enhanced priori skeleton structure[J]. Journal of ZheJiang University (Engineering Science), 2024, 58(1): 50-60.

Viewed

Full text

Abstract

Cited

Shared

Discussed