基于点云数据的黄土公路边坡病害识别  

Disease Detection of Loess Highway Slope Based on Point Cloud Data

作　　者：邵建鸿 郑万鹏 张斌 杜源 王兴涛 栾纪昊 Shao Jianhong;Zheng Wanpeng;Zhang Bin;Du Yuan;Wang Xingtao;Luan Jihao(GanSu Provincial Highway Development Group Co.,Ltd.,Lanzhou 730000,Gansu,China;Gansu Erche Highway Project Man-agement Co.,Ltd.,Qingyang 745000,Gansu,China;Gansu Henglu Communication Survey and Design Institute Co.,Ltd.,Lanzhou 730000,Gansu,China;Gansu Engineering Research Centre of Road Maintenance Technologies,Lanzhou 730000,Gansu,China)

机构地区：[1]甘肃省公路发展集团有限公司,甘肃兰州730000 [2]甘肃二车高等级公路项目管理有限公司,甘肃庆阳745000 [3]甘肃恒路交通勘察设计院有限公司,甘肃兰州730000 [4]甘肃省高等级公路养护工程研究中心,甘肃兰州730000

出　　处：《岩土工程技术》2023年第3期320-326,共7页Geotechnical Engineering Technique

基　　金：甘肃省交通运输厅科研项目(No.2020-22)。

摘　　要：公路边坡病害巡查是高速公路安全运维的重要工作之一。鉴于当前人工巡查的局限性,提出了一种基于无人机倾斜摄影技术实现三维重建和点云数据分析的公路边坡坡面病害巡查方法。研究结果表明:无人机倾斜摄影可以快速获得研究区的实景影像和点云数据,通过三维模型重建和点云数据分析,能够高效定量地识别出坡面变形、冲沟以及排水沟淤堵等边坡病害;点云数据对比算法对边坡病害识别结果有较大影响,相比最邻近点云比较方法和基于法向量的点云比较方法,点云网格比较方法更适用于公路边坡坡面病害识别。该方法提高了边坡巡查的效率,弥补了人工巡查的不足。Slope inspection is one of the main works of highway operation and maintenance.In view of the limitation of manual inspection and monitoring,a highway slope inspection and monitoring method based on 3D reconstruction and point cloud analysis using unmanned aerial vehicle(UAV)-based oblique photography technique was proposed.The results show that the terrain data of research area could be obtained quickly by the unmanned aerial vehicle(UAV)-based oblique photography technique.Slope diseases including slope deformation,gully erosion and blockage of drainage ditch could be identified quantitatively by 3D model reconstruction and point cloud data analysis.The algorithm of point cloud data analysis has a great impact on the results.Compared with Cloud to Cloud comparison(C2C)algorithm and Multiscale Model to Model Cloud comparison(M3C2)algorithm,the algorithm of Cloud to Mesh comparison(C2M)is most suitable for the point cloud data analysis of highway slope disease identification.This method improves the efficiency of slope inspection and is an effective method to make up for the shortage of manual inspection.

关 键 词：道路工程 病害识别 点云数据分析 三维模型重建 边坡巡查 

分 类 号：U419.4[交通运输工程—道路与铁道工程]