摘要
道路边界精确提取建模是城市道路管理、智能交通规划和高精度地图制作等领域的重要课题之一。本文提出了一种基于车载激光雷达点云数据和开源街道地图(OSM)的三维道路边界精确提取方法。首先,针对原始车载LiDAR点云数据应用布料模拟滤波分离地面点,再结合相对高程分析获取道路边界点候选数据集。然后,应用OSM矢量道路网数据的节点辅助道路边界点候选点集进行分段。最后,在各分段点云数据集中基于随机抽样一致性算法获得三维道路边界点集。通过直道、弯道及高密度复杂场景3种不同类型的城区道路边界路段分类提取试验。结果表明,利用该方法进行道路边界提取的准确率和召回率分别达96.12%和95.17%,F1值达92.11%,本文方法可用于高精度道路边界的三维精细提取与矢量化,进而为智能交通与无人驾驶导航提供支撑。
Accurate road boundary extraction modeling is an important topic in urban road management,intelligent traffic planning and high-precision map generation.In this paper,an accurate 3D road boundary extraction method based on OSM is proposed based on mobile laser scanning point clouds data.Firstly,the original mobile LiDAR point cloud data is processed by CSF filtering to separate the ground points,and the candidate data set of road boundary points is obtained by combining with the relative elevation analysis.Then,the nodes of OSM vector road network data are used to assist the data segmentation of road boundary point candidate point set.Finally,a 3D road boundary point set is obtained based on RANSAC algorithm in each segment point cloud data set.Through the extraction experiment of three different types of urban road boundary sections,the analysis results show that,the accuracy rate and recall rate of the proposed method are 96.12%and 95.17%,respectively,and F1 value is 92.11%.The research method in this paper can be used to extract and vectorize high-precision road boundaries,thus provide support for intelligent transportation and unmanned navigation.
作者
王艳军
林云浩
王书涵
李少春
王孟杰
WANG Yanjun;LIN Yunhao;WANG Shuhan;LI Shaochun;WANG Mengjie(School of Mapping and Remote Sensing,Hunan University of Science and Technology,Hunan Provincial Key Laboratory of Geo-Information Engineering in Surveying,Xiangtan 411201,China;National-local Joint Engineering Laboratory of Geo-spatial Information Technology,Hunan University of Science and Technology,Xiangtan 411201,China;School of Resource Environment and Safety Engineering,Hunan University of Science and Technology,Xiangtan 411201,China)
出处
《测绘通报》
CSCD
北大核心
2022年第7期18-25,共8页
Bulletin of Surveying and Mapping
基金
国家自然科学基金(41971423,31972951,41771462)
湖南省自然科学基金(2020JJ3020)
湖南省科技计划项目(2019RS2043,2019GK2132)
湖南省教育厅优秀青年项目(18B224)。
