土木与环境工程学报（中英文）2025，Vol.47Issue(5)：86-96,11.DOI:10.11835/j.issn.2096-6717.2023.148
基于无人机图像与深度学习的高原地区隧道洞门墙病害检测方法
Disease detection method of tunnel portals in plateau region based on UAV images and deep learning
摘要
Abstract
In light of the prevalent diseases of tunnel portals built in the harsh environment of the plateau region and the problems of low efficiency and high risk associated with traditional manual disease detection methods,a novel disease detection method for tunnel portals in the plateau region based on Unmanned Aerial Vehicle(UAV)image and deep learning was proposed.Firstly,an UAV was used to collect the disease images of a tunnel portal in the plateau region of Xinjiang,and a multi-disease semantic segmentation dataset was constructed.Then,based on DeeplabV3+,an improved model TP-DeeplabV3+was proposed,which used MobileNetV2 as the backbone feature extraction network to reduce model parameters;Used Focal Loss as the loss function to solve the category imbalance problem in disease images;Used the CA attention mechanism to improve the segmentation performance;and proposed the disease quantification method.Experiment results show that TP-DeeplabV3+attains 88.37%and 94.93%of mIoU and mPA on the test set,respectively.Furthermore,the model volume is reduced by 88.83%.The absolute error of the proposed disease quantification method for disease coverage rate is less than 0.3%,and the relative error is maintained below 7.31%.Compared with the traditional manual method,the proposed method facilitates the intelligent detection of tunnel portal safely and accurately in plateau region.关键词
隧道洞门墙病害/深度学习/语义分割/无人机/高原地区Key words
tunnel portal disease/deep learning/semantic segmentation/unmanned aerial vehicle(UAV)/plateau regions分类
交通工程引用本文复制引用
车博文,包卫星,郭强,潘振华,卢汉青,尹严..基于无人机图像与深度学习的高原地区隧道洞门墙病害检测方法[J].土木与环境工程学报（中英文）,2025,47(5):86-96,11.基金项目
新疆重大科技专项(2020A03003-7) （2020A03003-7）
陕西省自然科学基础研究计划(2021JM-180) （2021JM-180）
中央高校基本科研业务费(领军人才计划)(300102211302)Major Science and Technology Projects in Xinjiang(No.2020A03003-7) （领军人才计划）
Shaanxi Province Natural Science Basic Research Project(No.2021JM-180) （No.2021JM-180）
Basic Scientific Research Funds of Central Universities(Leading Talents Program)(No.300102211302) （Leading Talents Program）
