基于深度学习的隧道病害图像检测  

Image Detection of Disease in Cross-river Tunnel Based on Deep Learning

作　　者：高新闻[1,2] 王龙坤 GAO Xinwen;WANG Longkun(School of Mechanical and Electrical Engineering and Automation,Shanghai University,Shanghai 200444,China;SHU-SUCG Research Centre of Building Information,Shanghai University,Shanghai 201400,China)

机构地区：[1]上海大学机电工程与自动化学院,上海200444 [2]上海大学上海城建(集团)公司建筑产业化研究中心,上海201400

出　　处：《计算机测量与控制》2022年第2期58-64,共7页Computer Measurement &Control

摘　　要：随着我国城市地铁的快速发展,隧道的养护变得越来越重要,传统的人工检测方法不仅效率低、成本高,而且耗时,已经不能满足当今的需求;通过对越江隧道中的电缆通道的病害特征进行研究,提出一种基于深度学习的隧道多病害检测的方法,并提出了一种针对隧道病害检测的残差融合模块网络(Resfmnet),利用深度学习网络提取图像病害特征并进行病害分类,提高了病害的检测能力,所使用的数据集是通过特种机器人在越江隧道中的电缆通道拍摄的视频获得;实验结果表明所提出的网络显示出更高的准确性和泛化性,对多病害的检测的精度mAP达到0.8914,使得越江隧道检查和监控变得高效、低成本,并最终实现自动化。With the rapid development of urban subways in our country,the maintenance of tunnels becomes more and more important.The traditional manual inspection methods are not only low in efficiency and high in cost,but also time-consuming,which can no longer meet today's needs.By studying the disease characteristics of the cable channel in the cross-river tunnel,a method for detecting multiple diseases in the tunnel based on deep learning is proposed,and a residual fusion module network(Resfmnet)for the detection of tunnel diseases is proposed,using deep learning the network extracts image disease features and performs disease classification,which improves the detection ability of diseases.In the cable channel of the cross-river tunnel,the data set is used to be obtained from the video taken by special robots;The experimental results show that the proposed network has higher accuracy and generalization,and the accuracy mAP of multi-disease detection reaches 0.8914,which makes the inspection and monitoring of the cross-river tunnel more efficient and low-cost,and finally realizes automation.

关 键 词：越江隧道 电缆通道 隧道病害检测 深度学习 自动化 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]