地铁隧道结构机器视觉检测系统及应用分析

doi:10.13474/j.cnki.11-2246.2020.0277

测绘通报 ›› 2020, Vol. 0 ›› Issue (9): 27-32,37.doi: 10.13474/j.cnki.11-2246.2020.0277

• 轨道交通前沿测绘技术研究与应用 • 上一篇下一篇

地铁隧道结构机器视觉检测系统及应用分析

李军¹, 朱国琦¹, 樊晓东², 杨维², 黄震³

1. 南宁轨道交通集团有限责任公司, 广西南宁 530029;
2. 宽衍(北京)科技发展有限公司, 北京 100089;
3. 广西大学土木建筑工程学院, 广西南宁 530004

收稿日期:2020-06-15 修回日期:2020-07-16 出版日期:2020-09-25 发布日期:2020-09-28
通讯作者: 樊晓东。E-mail:619631361@qq.com E-mail:619631361@qq.com
作者简介:李军(1980-),男,高级工程师,主要从事城市轨道交通运营管理工作。E-mail:83666545@qq.com

Machine vision inspection system of subway tunnel structure and its application analysis

LI Jun¹, ZHU Guoqi¹, FAN Xiaodong², YANG Wei², HUANG Zhen³

1. Nanning Rail Transit Group Co., Ltd., Nanning 530029, China;
2. Kuanyan(Beijing) Technology Development Co., Ltd., Beijing 100089, China;
3. College of Civil Engineering and Architecture, Guangxi University, Nanning 530004, China

Received:2020-06-15 Revised:2020-07-16 Online:2020-09-25 Published:2020-09-28

摘要/Abstract

摘要： 本文使用一种地铁隧道快速机器视觉检测系统，该系统采用深度学习算法，有效识别了采集图像中的病害特征。其软件平台能够方便隧道管理者实时查看和跟踪隧道病害信息。该检测系统应用于南宁地铁的检测，取得了较好的检测效果，能精确地识别隧道的裂缝、剥落和渗漏等病害，并从图像效果、定位精度、检测效率、识别率和检测精度4个方面与人工检测方法进行了对比分析。结果表明，该系统的图像采集效果显著优于人工方法；检测速度可达30 km/h；识别裂缝、剥落和渗漏的识别精度分别可达89%、100%和94%；能够识别面积为100 mm²的表面缺陷和裂缝宽度>0.2 mm的裂缝。

关键词: 地铁隧道, 机器视觉, 图像处理, 快速检测, 应用分析

Abstract: This paper presents a rapid machine vision detection system for subway tunnel. The system uses deep learning algorithm to effectively identify the disease features in the collected images. The software platform can facilitate tunnel managers to view and track tunnel disease information in real time. The detection system has been applied to the detection of Nanning metro, and has achieved good detection results. It can accurately identify the tunnel defects such as cracks, spalling and leakage. It also compares with the manual detection method in terms of image effect, positioning accuracy, detection efficiency, recognition rate and detection accuracy. The results show that the image acquisition effect of the system is significantly better than that of the manual method; the detection speed can reach 30 km/h; the recognition accuracy of crack, spalling and leakage can reach 89%, 100% and 94% respectively; the surface defects with an area of 100 mm² and cracks with a width of more than 0.2 mm can be identified.

Key words: subway tunnel, machine vision, image processing, rapid detection, application analysis

中图分类号:

P258

李军, 朱国琦, 樊晓东, 杨维, 黄震. 地铁隧道结构机器视觉检测系统及应用分析[J]. 测绘通报, 2020, 0(9): 27-32,37.

LI Jun, ZHU Guoqi, FAN Xiaodong, YANG Wei, HUANG Zhen. Machine vision inspection system of subway tunnel structure and its application analysis[J]. Bulletin of Surveying and Mapping, 2020, 0(9): 27-32,37.

参考文献

[1] 傅鹤林,黄震, 黄宏伟,等. 基于云理论的隧道结构健康诊断方法[J]. 工程科学学报,2017,39(5):794-801.
[2] 宋云记,王智. 利用三维激光扫描技术进行地铁隧道施工质量管控及病害检测[J]. 测绘通报,2020(5):150-154.
[3] 傅鹤林,黄震,王慧,等. 地铁安全事故分析及安全管理[J]. 隧道与地下工程灾害防治,2019,1(2):59-66.
[4] HUANG Z, FU H L, ZHANG J B, et al. Structural damage evaluation method for metro shield tunnel[J]. Journal of Performance of Constructed Facilities, 2019, 33(1):04018097.
[5] HUANG H W, SUN Y, XUE Y D, et al. Inspection equipment study for subway tunnel defects by grey-scale image processing[J].Advanced Engineering Informatics, 2017, 32:188-201.
[6] 高洪,李凯,马全明,等. 移动三维激光测量系统在地铁运营隧道病害监测中的应用研究[J]. 测绘通报,2019(8):96-101,161.
[7] SASAMA H, UKAI M, OHTA M, et al. Inspection system for railway facilities using a continuously scanned image[J]. Electrical Engineering in Japan, 1998,125 (2):52-64.[8] NISHIYAMA S, MINAKATA N, KIKUCHI T, et al. Improved digital photogrammetry technique for crack monitoring[J]. Advanced Engineering Informatics, 2015,29(4):851-858.[9] MENENDEZ E, VICTORES J G, MONTERO R, et al. Tunnel structural inspection and assessment using an autonomous robotic system[J]. Automation in Construction, 2018, 87:117-126.[10] 杨俊,刘笑娣,刘新根,等. 公路隧道结构快速检测车综述[J]. 华东交通大学学报,2018,35(4):30-38.[11] 王平让. 隧道病害快速检测技术现状及分析比较[J]. 公路与汽运,2016(3):241-245.[12] CAI Z, VASCONCELOS N. Cascade R-CNN:delving into high quality object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT:IEEE, 2018:6154-6162.[13] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553):436-444.[14] XIE S, GIRSHICK R, DOLLAR P, et al. Aggregated Residual Transformations for Deep Neural Networks[C]//IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, HI:IEEE, 2017:5987-5995.[15] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(2):318-327.[16] CHENG Chuanqi, HAO Xiangyang, LI Jiansheng,et al. A robust Gaussian mixture model for mobile robots' vision-based pose estimation[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(3):79-90.[17] SHANG Yang,X SUN iaoliang, ZHANG Yueqiang,et al. Research on 3D target pose tracking and modeling[J]. Journal of Geodesy and Geoinformation Science, 2019, 2(2):60-69.

地铁隧道结构机器视觉检测系统及应用分析

Machine vision inspection system of subway tunnel structure and its application analysis

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	时梦杰, 陶庭叶, 高飞, 李江洋, 陈皓. 雾天环境下机器视觉的边坡监测方法[J]. 测绘通报, 2022, 0(6): 88-92.
[2]	李旭. 三维激光扫描技术在地铁隧道竣工测量中的应用[J]. 测绘通报, 2022, 0(6): 166-169.
[3]	潘东峰, 杨超, 吴一同, 谭钿, 张德林. 利用TLS技术进行地铁隧道断面提取及变形监测分析[J]. 测绘通报, 2022, 0(4): 130-133.
[4]	贾永红. 数字图像处理混合教学的研究与实践[J]. 测绘通报, 2022, 0(2): 174-176.
[5]	程瑞, 高建, 邢强, 孙中昶. 一种优化的Faster R-CNN小目标检测方法[J]. 测绘通报, 2021, 0(9): 21-27.
[6]	丁孝兵, 高志强, 杨坤. 基于惯导和CPⅢ控制点的地铁隧道移动激光扫描三维点云重建[J]. 测绘通报, 2021, 0(9): 112-115,129.
[7]	董之南, 时培好, 高晗. 基于移动式激光扫描的点云数据处理[J]. 测绘通报, 2021, 0(8): 97-101.
[8]	陈向阳, 侯刚栋, 潘恺, 李成成. 基于车载式激光扫描的地铁隧道结构检测[J]. 测绘通报, 2021, 0(1): 165-168.
[9]	赵亚波, 王智. 基于移动三维扫描技术的隧道管片错台分析及应用[J]. 测绘通报, 2020, 0(8): 160-163.
[10]	宋云记, 王智. 利用三维激光扫描技术进行地铁隧道施工质量管控及病害检测[J]. 测绘通报, 2020, 0(5): 150-154.
[11]	李航, 马焱, 张翔. 一种网络地图风格化滤镜与中间件技术[J]. 测绘通报, 2020, 0(11): 85-89.
[12]	高洪, 李凯, 马全明, 韩志晟, 孙丕川. 移动三维激光测量系统在地铁运营隧道病害监测中的应用研究[J]. 测绘通报, 2019, 0(8): 96-101,161.
[13]	严成, 何宁, 庞维庆, 邓德迎. 基于机器视觉的水域远程监测应用系统方案设计[J]. 测绘通报, 2019, 0(8): 149-153.
[14]	贾永红, 付仲良, 孙和利. 中国大学MOOC平台数字图像处理在线开放课程的建设与应用[J]. 测绘通报, 2018, 0(12): 144-147.
[15]	张嵘. 基于Matlab平台的遥感图像变化检测算法改进策略[J]. 测绘通报, 2016, 0(7): 84-89.