基于三维激光技术的路面坑槽检测误差补偿算法研究  

Study on Error Compensation Algorithm for Pavement Pothole Detection Based on 3D Laser Technology

作　　者：周毅姝 张金凝 曹瑾瑾 刘越 张浩然[2] ZHOU Yi-shu;ZHANG Jin-ning;CAO Jin-jin;LIU Yue;ZHANG Hao-ran(Research Institute of Highway,Ministry of Transport,Beijing 100088,China;School of Transportation,Southeast University,Nanjing Jiangsu 211189,China)

机构地区：[1]交通运输部公路科学研究院,北京100088 [2]东南大学交通学院,江苏南京211189

出　　处：《公路交通科技》2023年第3期17-24,共8页Journal of Highway and Transportation Research and Development

基　　金：中央级公益性科研院所基本科研业务费专项资金项目(2021-9044b)。

摘　　要：车载式三维激光设备在道路工程路面检测过程中因受沥青混合料表面纹理及检测条件等因素的影响,导致检测结果的准确度较低。为了提高该设备的检测精度,模拟了实际沥青路面坑槽形态,设计加工了2个坑槽模型用于试验研究。以高精度手持式三维激光设备获取的坑槽点云数据作为基准,与车载式三维激光设备检测的坑槽数据进行了对比,分析了坑槽深度、面积和体积3个指标。基于三维坐标变换原理建立了坑槽点云数据修正模型,提出了一种基于等高线提取的坑槽多维度指标误差补偿算法。通过提取坑槽等高线轮廓点云数据,计算出坑槽等高线轮廓面积,得出了坑槽等高线与等高线轮廓面积的关系,并计算出点云数据的坐标修正量,根据坑槽点云数据修正模型对坑槽点云数据进行了修正。结果表明:通过回归分析的方法可得到坑槽等高线轮廓面积与等高线的关系;坑槽模型1的面积和体积误差修正前为3.2%和3.4%,修正后为2.2%和1.1%;坑槽模型2的面积和体积误差修正前为13.0%和4.6%,修正后为2.9%和1.4%;该算法应用于修正车载式三维激光设备获取的沥青路面坑槽点云数据后,沥青路面坑槽面积和体积的误差分别由-3.4%和-8.3%修正为0.5%和0.7%。研究成果验证了该补偿算法可应用于修正车载式三维激光设备的检测数据,提高沥青路面坑槽的检测精度,可为路面坑槽严重程度评估的可靠度提供保障。The vehicle-mounted 3D laser equipment is affected by the factors such as the surface texture of asphalt mixture and the detection conditions during highway engineering pavement detection process,leading in inaccurate detection results.In order to improve the detection accuracy of the equipment,the pothole shape in actual asphalt pavement is simulated,and 2 pothole models are designed and produced for experimental research.Using the pothole pointcloud data obtained by the high-precision handheld 3D laser equipment as the benchmark,the pothole data detected by the vehicle-mounted 3D laser equipment are compared,and the depth,area and volume of the pothole are analyzed.Based on the principle of 3D coordinate transformation,a model for correction of pothole pointcloud data is established,and a pothole multi-dimensional indicator error compensation algorithm based on contour line extraction is proposed.By extracting the contour pointcloud data from the contour line of the pothole,the contour area of the contour line of the pothole is calculated,and the relationship between the contour line and the contour area of the pothole is obtained.The coordinate correction amounts of the pointcloud data are calculated,and the pothole pointcloud data are corrected according to the established model.The result shows that(1)the relationship between the pothole contour line contour area and the contour line could be obtained by the regression analysis method;(2)the area and volume errors of the pothole model 1 are 3.2%and 3.4%before correction,2.2%and 1.1%after correction;(3)the area and volume errors of pothole model 2 are 13.0%and 4.6%before correction,2.9%and 1.4%after correction;(4)after applying the algorithm to correct the pothole pointcloud data of asphalt pavement obtained by vehicle-mounted 3D laser equipment,the errors of pothole area and volume of asphalt pavement are corrected from-3.4%and-8.3%to 0.5%and 0.7%respectively.The research result verified that the compensation algorithm could be applied to correct the detecti

关 键 词：道路工程 误差补偿 多维度指标 坑槽 等高线 

分 类 号：U416.03[交通运输工程—道路与铁道工程]