基于DenseNet分类的隧道裂缝检测研究  

Study on Tunnel Crack Detection Based on DenseNet Classification

作　　者：高新闻[1,2] 李帅青 金邦洋 Gao Xinwen;Li Shuaiqing;Jin Bangyang(Institute of Mechanical and Electrical Engincering and Automation,ShanghaiUniversity,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》2020年第8期58-61,87,共5页Computer Measurement &Control

基　　金：上海市科技委员会项目(17DZ1204203);上海市科技委员会项目(18DZ1201204)。

摘　　要：针对隧道裂缝人工识别低效、检修不便以及隧道环境复杂多变、检测易受噪声干扰等问题,文中提出一种基于深度学习的裂缝检测算法;通过神经网络对原始图像进行非裂缝区域过滤,减少无关背景信息的干扰,同时在分割算法基础上通过多维分类器将误识别的裂缝区域剔除;实验结果表明,密集连接卷积网络(DenseNet)在裂缝分类中最高可达99.95%的准确率,有效提升了隧道裂缝自动检测精度。Aiming at the problems of inefficient manual detection of tunnel cracks,inconvenient maintenance,complicated and changeable tunnel environment,and susceptibility to noise interference,a crack detection algorithm based on deep learning is proposed.Non-crack areas are filtered through the neural network to reduce the interference of irrelevant background information.At the same time,based on the segmentation algorithm,mis-recognized crack areas are eliminated.Experimental results show that the DenseNet network can reach a maximum accuracy of 99.95%in crack classification,which effectively improves the accuracy of tunnel crack detection.

关 键 词：裂缝检测 深度学习 DenseNet 

分 类 号：TP934[自动化与计算机技术]