基于深度语义分割的隧道渗漏水智能识别

doi:10.13722/j.cnki.jrme.2022.0016

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (2042 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要为解决现有隧道渗漏水检测方法中存在的检测精度较低、抗干扰能力较差、检测速度较慢的问题，提出了一种基于深度语义分割的隧道渗漏水图像识别算法。该算法以DeepLabv3+语义分割算法为基础，首先采用轻量化分类网络EfficientNetv2作为主干网络，在减少网络参数的同时，提升了识别精度；其次融合卷积注意力机制模块，通过增大图像中有效通道的权重，提升网络对于渗漏水特征信息的提取能力；进而从图像识别精度、模型内存和处理速度3个方面与DeepLabv3+，PSPnet，Unet等传统语义分割算法进行对比实验。研究表明：提出算法的平均像素准确率和平均交并比分别为93.99%，89.87%，模型大小仅为33.4 MB，图像处理速度可达39.87 f/s。相较于3种对比算法，构建算法在隧道渗漏水病害检测的精度和效率上都有显著提升，且具有更为优越的边缘分割效果以及抗干扰能力，适用于复杂环境下的隧道渗漏水检测任务，能够更好地满足工程检测需求。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	周中1
	2
	张俊杰1
	龚琛杰1
	2
	丁昊晖1

关键词 ：隧道工程, 渗漏水, 语义分割, 深度学习, 神经网络

Abstract：Aiming to solve the challenges of low detection accuracy，poor anti-interference ability and slow detection speed in the traditional tunnel leakage detection methods，a depth semantic segmentation algorithm for tunnel leakage is proposed on the basis of the DeepLabv3+ semantic segmentation algorithm. Firstly，the lightweight classification network EfficientNetv2 is used as the backbone network，which enhances the recognition accuracy while reduces network parameters. Secondly，Convolutional Block Attention Module(CBAM) is integrated to increase the weight of the effective channels in the image，thereby improving the ability of extraction of leakage feature information. Traditional semantic segmentation algorithms，including DeepLabv3+，PSPnet and Unet，are used for comparative experiments from three aspects：image recognition accuracy，model size，and detection speed. The results show that the mean pixel accuracy(mPA)，mean intersection over union(mIoU)，model size and image processing speed(FPS) of the proposed algorithm are 93.99%，89.87%，33.4 MB and 39.87 f/s，respectively. Compared with the three comparison algorithms，the detection accuracy and efficiency of the proposed algorithm have been significantly improved. Furthermore，the proposed algorithm has better edge segmentation effect and anti-interference ability，which is suitable for tunnel leakage detection tasks in complex environments，so as to better meet the needs of engineering detection.

Key words： tunnelling engineering leakage semantic segmentation deep learning neural network

引用本文:

周中1，2，张俊杰1，龚琛杰1，2，丁昊晖1. 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 Zhong1，2，ZHANG Junjie1，GONG Chenjie1，2，DING Haohui1. Automatic identification of tunnel leakage based on deep semantic segmentation . , 2022, 41(10): 2082-2093.

链接本文:

http://rockmech.whrsm.ac.cn/CN/10.13722/j.cnki.jrme.2022.0016 或 http://rockmech.whrsm.ac.cn/CN/Y2022/V41/I10/2082

[43]	TAN M，LE Q. Efficientnetv2：Smaller models and faster training[C]// International Conference on Machine Learning. New York：PMLR，2021：10 096–10 106.
[28]	REDMON J，FARHADI A. YOLOv3：an incremental improvement[J]. arXiv preprint arXiv，2018，http：//doi.org/10.48550/arXiv.1804.02767.
[6]	WEI F，YAO G，YANG Y，et al. Instance-level recognition and quantification for concrete surface bughole based on deep learning[J]. Automation in Construction，2019，107：102920.
[16]	刘学增，桑运龙，苏云帆. 基于数字图像处理的隧道渗漏水病害检测技术[J]. 岩石力学与工程学报，2012，31(增2)：3 779–3 786. (LIU Xuezeng，SANG Yunlong，SU Yunfan. Tunnel leakage disease detection technology based on digital image processing[J]. Chinese Journal of Rock Mechanics and Engineering，2012，31(Supp.2)：3 779–3 786.(in Chinese))
[18]	KRIZHEVSKY A，SUTSKEVER I，HINTON G E. Imagenet classification with deep convolutional neural networks[C]// Proceedings of the 25th International Conference on Neural Information Processing Systems. New York：Curran Associates Inc，2012：1–9.
[30]	GIRSHICK R. Fast R-CNN[C]// Proceedings of the IEEE International Conference on Computer Vision. Piscataway：IEEE，2015：1 440–1 448.
[40]	YU F，KOLTUN V. Multi-scale context aggregation by dilated convolutions[J]. arXiv preprint arXiv，2015，https://doi.org/10.48550/ arXiv.1511.07122.
[36]	黄宏伟，李庆桐. 基于深度学习的盾构隧道渗漏水病害图像识别[J]. 岩石力学与工程学报，2017，36(12)：2 861–2 871.(HUANG Hongwei， LI Qingtong. Image recognition for water leakage in shield tunnel based on deep learning[J]. Chinese Journal of Rock Mechanics and Engineering，2017，36(12)：2 861–2 871.(in Chinese))
[41]	CHEN L C，PAPANDREOU G，KOKKINOS I，et al. Semantic image segmentation with deep convolutional nets and fully connected CRFs[J]. arXiv preprint arXiv，2014，https://doi.org/10.48550/ arXiv.1412.7062.
[44]	WOO S，PARK J，LEE J Y，et al. Cbam：Convolutional block attention module[C]// Proceedings of the European conference on computer vision(ECCV). Berlin：Springer，2018：3–19.
[46]	RUSSELL B C，TORRALBA A，MURPHY K P，et al. LabelMe：a database and web-based tool for image annotation[J]. International Journal of Computer Vision，2008，77(1)：157–173.
[1]	ZHOU Z，DING H，MIAO L，et al. Predictive model for the surface settlement caused by the excavation of twin tunnels[J]. Tunnelling and Underground Space Technology，2021，114：104014.
[4]	GONG C，DING W，SOGA K，et al. Failure mechanism of joint waterproofing in precast segmental tunnel linings[J]. Tunnelling and Underground Space Technology，2019，84：334–352.
[11]	OTSU N. A threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems，Man，and Cybernetics，1979，9(1)：62–66.
[14]	ALEKSEYCHUK O. Detection of crack-like indications in digital radiography by global optimisation of a probabilistic estimation function[Ph. D. Thesis][D]. Zurich：Bundesanstalt für Materialforschung und-prüfung(BAM)，2006.
[21]	HE K，ZHANG X，REN S，et al. Deep residual learning for image recognition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway：IEEE，2016：770–778.
[24]	SILVA W R L，LUCENA D S. Concrete cracks detection based on deep learning image classification[C]// Proceedings of the Multidisciplinary Digital Publishing Institute. Basel：MDPI，2018，2(8)：489.
[31]	REN S，HE K，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，2016，39(6)：1 137–1 149.
[34]	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.
[7]	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.
[9]	刘宇飞，樊健生，聂建国，等. 结构表面裂缝数字图像法识别研究综述与前景展望[J]. 土木工程学报，2021，54(6)：79–98.(LIU Yufei，FAN Jiansheng，NIE Jianguo，et al. Review and prospect of digital-image-based crack detection of structure surface[J]. China Civil Engineering Journal，2021，54(6)：79–98.(in Chinese))
[17]	彭斌，祝志恒，阳军生，等. 基于全景展开图像的隧道衬砌渗漏水数字化识别方法研究[J]. 现代隧道技术，2019，56(3)：31–37.(PENG Bin，ZHU Zhiheng，YANG Junsheng，et al. On digital identification of water leakage at tunnel lining based on the panoramic developed image[J]. Modern Tunnel Technology，2019，56(3)：31–37.(in Chinese))
[19]	SZEGEDY C，LIU W，JIA Y，et al. Going deeper with convolutions[C]// Proceedings of the IEEE Conference On Computer Vision and Pattern Recognition. Piscataway：IEEE，2015：1–9.
[27]	REDMON J，DIVVALA S，GIRSHICK R，et al. You only look once：unified，real-time object detection[C]// Computer Vision and Pattern Recognition. Las Vegas，USA：IEEE，2016：779–788.
[8]	鲍跃全，李惠. 人工智能时代的土木工程[J]. 土木工程学报，2019，52(5)：1–11.(BAO Yuequan，LI Hui. Artificial Intelligence for civil engineering[J] China Civil Engineering Journal，2019，52(5)：1–11.(in Chinese))
[29]	GIRSHICK R，DONAHUE J，DARRELL T，et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// IEEE Conference on Computer Vision and Pattern Recognition. Los Alamitos，CA：IEEE Computer Society，2014：580–587.
[37]	ZHANG L，SHEN J，ZHU B. A research on an improved Unet-based concrete crack detection algorithm[J]. Structural Health Monitoring，2021，20(4)：1 864–1 879.
[39]	CHOLLET F. Xception：Deep learning with depthwise separable convolutions[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway：IEEE，2017：1 251–1 258.
[26]	REDMON J，FARHADI A. YOLO9000：better，faster，stronger[C]// IEEE Conference on Computer Vision and Pattern Recognition. Hawaii，USA：IEEE，2017：6 517–6 525.
[47]	LIN T Y，GOYAL P，GIRSHICK R，et al. Focal loss for dense object detection[C]// Proceedings of the IEEE International Conference on Computer Vision. Piscataway：IEEE，2017：2 980–2 988.
[2]	LEI M，ZHU B，GONG C，et al. Sealing performance of a precast tunnel gasketed joint under high hydrostatic pressures：Site investigation and detailed numerical modeling[J]. Tunnelling and Underground Space Technology，2021，115：104082.
[3]	GONG C，DING W，XIE D. Parametric investigation on the sealant behavior of tunnel segmental joints under water pressurization[J]. Tunnelling and Underground Space Technology，2020，97：103231.
[25]	LIU W，ANGUELOV D，ERHAN D，et al. Ssd：Single shotmultibox detector[C]// European conference on computer vision. Cham：Springer，2016：21–37.
[5]	ZHANG J，LI F，ZENG L，et al. Effect of cushion and cover on moisture distribution in clay embankments in southern China[J]. Journal of Central South University，2020，27(7)：1 893–1 906.
[10]	WEI F，YAO G，YANG Y，et al. Instance-level recognition and quantification for concrete surface bughole based on deep learning[J]. Automation in Construction，2019，107：102920.
[12]	马丽莎. 基于数字图像处理的路面裂缝识别方法研究[硕士学位论文][D]. 南京：东南大学，2018.(MA Lisha Research on pavement crack recognition method based on digital image processing[M. S. Thesis][D]. Nanjing：Southeast University，2018.(in Chinese))
[13]	LI Q，ZOU Q，ZHANG D，et al. FoSA：F* seed-growing approach for crack-line detection from pavement images[J]. Image and Vision Computing，2011，29(12)：861–872.
[15]	FUJITA Y，MITANI Y，HAMAMOTO Y. A method for crack detection on a concrete structure[C]//18th International Conference on Pattern Recognition (ICPR'06). Piscataway：IEEE，2006：901–904.
[20]	SIMONYAN K，ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[C]// International Conference on Learning Representations. San Diego：ICLR，2015：1–14..
[22]	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.(in Chinese))
[23]	WANG W，HU W，WANG W，et al. Automated crack severity level detection and classification for ballastless track slab using deep convolutional neural network[J]. Automation in Construction，2021，124：103484.
[32]	任松，朱倩雯，涂歆玥，等. 基于深度学习的公路隧道衬砌病害识别方法[J]. 浙江大学学报：工学版，2022，56(1)：92–99.(REN Song，ZHU Qingwen，TU Xinyue，et al. Lining disease identification of highway tunnel based on deep learning[J]. Journal of ZheJiang University：Engineering Science，56(1)：92–99.(in Chinese))
[33]	DENG J，LU Y，LEE V C S. Concrete crack detection with handwriting script interferences using faster region-based convolutional neural network[J]. Computer-Aided Civil and Infrastructure Engineering，2020，35(4)：373–388.
[35]	FENG C，ZHANG H，LI Y，et al. Efficient real-time defect detection for spillway tunnel using deep learning[J]. Journal of Real-Time Image Processing，2021，18(6)：1–11.
[38]	CHEN L，ZHU Y，PAPANDREOU G，et al. Encoder-decoder with atrous separable convolution for semantic image segmentation[C]// European Conference on Computer Vision. Berlin，German：Springer，2018：801–818.
[42]	CHEN L C，PAPANDREOU G，SCHROFF F，et al. Rethinking atrous convolution for semantic image segmentation[J]. arXiv preprint arXiv，2017，https://doi.org/10.48550/arXiv.1706.05587.
[45]	TAN M，LE Q. Efficientnet：Rethinking model scaling for convolutional neural networks[C]// International Conference on Machine Learning. New York：PMLR，2019：6 105–6 114.
[48]	LIASHCHYNSKYI P，LIASHCHYNSKYI P. Grid search，random search，genetic algorithm：A big comparison for NAS[J]. arXiv preprint arXiv，2019，https://doi.org/10.48550/arXiv.1912.06059.