Improved Mask RCNN method for shield tunnel leakage detection

doi:10.13474/j.cnki.11-2246.2024.0232

Abstract

Abstract: Water leakage is an important characterization of the potential damage or defects of the shield tunnel structure. Rapid and accurate detection of the tunnel leakage site is of great significance to the safe operation and maintenance of the tunnel. However, most of the existing methods use optical images to detect the tunnel water leakage, but due to the tunnel space limitations and light conditions, it is difficult to obtain high-quality disease pictures. In this regard, a water leakage detection method based on terrestrial laser scanning point cloud and improved Mask RCNN is proposed. Firstly, the laser point cloud reflection intensity is corrected, and then the gray scale image is generated and the water leakage disease data set is established. Finally, the atrous convolution and deformation convolution are introduced in the Mask RCNN algorithm to realize the rapid detection of tunnel water leakage disease. The data collected in metro are used for verification. Experimental results show that, compared to the original algorithm and FCN algorithm, the detection accuracy of the proposed improved Mask RCNN algorithm is significantly improved, and it has a good performance in water leakage identification in shield tunnel.

Key words: shield tunnel, point cloud, reflection intensity correction, Mask RCNN, water leakage detection

CLC Number:

P258

WANG Jian, ZHENG Like, WU Binjie, QI Zhiyu. Improved Mask RCNN method for shield tunnel leakage detection[J]. Bulletin of Surveying and Mapping, 2024, 0(2): 170-177.

References

[1] YUAN Yong,JIANG Xiaomo,LIU Xian. Predictive maintenance of shield tunnels[J]. Tunnelling and Underground Space Technology,2013,38: 69-86.
[2] 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.
[3] HUANG Hongwei,SUN Yan,XUE Yadong,et al. Inspection equipment study for subway tunnel defects by grey-scale image processing[J]. Advanced Engineering Informatics,2017,32: 188-201.
[4] 黄永杰,柳献,袁勇,等. 盾构隧道渗漏水的自动检测技术[J]. 上海交通大学学报,2012,46(1): 73-78.
[5] YU Haiyang,HU Yawen,GUO Hongyu,et al. Tunnel moving target detection based on local structure of image and gray scale information[C]//Proceedings of 2016 IEEE International Conference on Intelligent Transportation Engineering (ICITE).Singapore:IEEE,2016: 103-107.
[6] 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.
[7] CHA Y,CHOI W,SUH G,et al. Autonomous structural visual inspection using region‐based deep learning for detecting multiple damage types[J]. Computer-Aided Civil and Infrastructure Engineering,2018,33(9): 731-747.
[8] HUANG Hongwei,LI Qingtong,ZHANG Dongming. 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] GARGEES R,MORAGO B,PELAPUR R,et al. Incident-supporting visual cloud computing utilizing software-defined networking[J]. IEEE Transactions on Circuits and Systems for Video Technology,2017,27(1): 182-197.
[10] 李珵,卢小平,朱宁宁,等. 基于激光点云的隧道断面连续提取与形变分析方法[J]. 测绘学报,2015,44(9): 1056-1062.
[11] ZHU Xi,WANG Tiejun,DARVISHZADEH R,et al. 3D leaf water content mapping using terrestrial laser scanner backscatter intensity with radiometric correction[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2015,110: 14-23.
[12] GARROWAY K,HOPKINSON C,JAMIESON R. Surface moisture and vegetation influences on LiDAR intensity data in an agricultural watershed[J]. Canadian Journal of Remote Sensing,2011,37(3): 275-284.
[13] KAASALAINEN S,JAAKKOLA A,KAASALAINEN M,et al. Analysis of incidence angle and distance effects on terrestrial laser scanner intensity: search for correction methods[J]. Remote Sensing,2011,3(10): 2207-2221.
[14] KASHANI A,OLSEN M,PARRISH C,et al. A review of LiDAR radiometric processing:form ad hoc intensity correction to rigorous radiometric calibration[J].Sensors,2015,15(11): 28099-28128.
[15] HE Kaiming,ZHANG Xiangyu,REN Shaoqing,et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2015,37(9): 1904-1916.
[16] YANG Haiping,YU Bo,LUO Jiancheng,et al. Semantic segmentation of high spatial resolution images with deep neural networks[J].GIS Science & Remote Sensing,2019,56(5):749-768.
[17] CHEN Fuchen,JAHANSHAHI M R. NB-CNN: deep learning-based crack detection using convolutional neural network and Naïve Bayes data fusion[J]. IEEE Transactions on Industrial Electronics,2018,65(5): 4392-4400.
[18] 王晓静,唐超,杨晓飞. 激光点云在地铁盾构隧道病害诊断中的应用[J]. 测绘通报,2020(9): 33-37.
[19] 谭凯,程效军. 基于多项式模型的TLS激光强度值改正[J]. 中国激光,2015,42(3): 0314002.