基于车载激光LiDAR点云数据的路面坑槽自动提取方法研究  

Research on Automatic Extraction of Road Pits Based on Carborne LiDAR Point Cloud Data

作　　者：郑明丹 孙五斌 罗明生 ZHENG Mingdan;SUN Wubin;LUO Mingsheng(Zhejiang Provincial Institute of Surveying and Mapping Science and Technology,Hangzhou 311100,China)

机构地区：[1]浙江省测绘科学技术研究院,浙江杭州311100

出　　处：《测绘与空间地理信息》2024年第10期197-199,203,共4页Geomatics & Spatial Information Technology

摘　　要：以车载激光LiDAR扫描道路点云数据为研究对象,提出一种基于点云剖面特征描述的坑槽提取方法。首先,对原始点云数据进行滤波处理,获取地面点数据;其次,对道路横纵剖面进行道格拉斯-普克算法的轮廓拟合,将积分不变性与微分特征作为描述算法进行坑槽提取;最后,使用约束条件进行点云聚类实现噪声点的剔除,进一步识别确定提取坑槽。为了验证本文方法的有效性,使用某段道路点云数据进行实验,结果表明,本文方法能够有效提取得到道路面坑槽点,不受坑槽形状的限制,具有较高的精度。Taking the road point cloud data acquired by carborne LiDAR as the research object,a pit extraction method based on point cloud profile characteristic description is proposed.Firstly,the original point cloud data is filtered to obtain the surface point data;secondly,the contour fitting of Douglas Puck algorithm is carried out for the transverse and longitudinal sections of the road,and the integral invariance and differential characteristics are used as the description algorithm to extract the pits;finally,constraint conditions are used to cluster point clouds to eliminate noise points,further identify,determine and extracts pits.In order to verify the effectiveness of this method,we use a section of road point cloud data for experiments,and the results show that this method can effectively extract the road surface pit points,which is not limited by the pit shape,and has high accuracy.

关 键 词：车载激光LiDAR 点云滤波 坑槽 特征描述 

分 类 号：P232[天文地球—摄影测量与遥感]