摘要
针对道路标线三维矢量数据难以高效精确获取的问题,该文提出了一种从移动激光扫描数据中自动提取道路标线的新方法。基于平缓路面这个假设,利用邻域高程一致性的判断方法提取地面点。将地面点根据轨迹数据分割成多组剖面,对每个剖面上的点云进行强度直方图统计,找到强度值有突变的点。以此为种子点通过强度值区域生长以得到完整的标线,利用点云模板匹配的方法剔除错误点集。最后对标线点云进行矢量化得到三维矢量线。通过城市中大约2km长的移动激光点云数据的实验,证明本文提出的方法在提取道路标线方面能得到较好的结果。
Aiming at the problem that it is still difficult to obtain 3 Dvector data of road markings efficiently and accurately,this paper presented a novel method to automatically extract road markings from mobile laser scanning(MLS)point cloud data.Ground points were first filtered from raw input point clouds using neighborhood elevation consistency method.The basic assumption of the method was that the road surface was smooth.Points with small elevation-difference between neighborhoods were considered to be ground points.Then ground points were partitioned into a set of profiles according to trajectory data.The intensity histogram of points in each profile was generated to find intensity jumps in certain threshold which inversely to laser distance.The separated points were used as seed points to region grow based on intensity so as to obtain road mark of integrity,we used the point cloud template-matching method to refine the road marking candidates via removing the wrong point clusters.During experiment with a MLS point set of about 2 kilometers in a city center,our method provided a promising solution to the road markings extraction from MLS data.
作者
马浩
裴智惠
李婷婷
MA Hao;PEI Zhihui;LI Tingting(Chinese Academy of Surveying&Mapping,Beijing 100036,China;Beijing Geo-Vision Tech.Co.,Ltd.,Beijing 100070,China;Beijing CASRS Information Technology Co .,Ltd.,Beijing 100101,China;Fengxian Water Conservancy Bureau,Xuzhou,Jiangsu 221700,China)
出处
《测绘科学》
CSCD
北大核心
2019年第6期217-221,共5页
Science of Surveying and Mapping
