摘要
Abstract
This paper proposed a road boundary point extraction method based on mobile vehicle laser scanning point cloud data to address the problems of poor efficiency and low accuracy in obtaining road boundary information in the past.Firstly,in order to reduce the amount of original road point cloud data and improve the efficiency of subsequent processing algorithms,a Volex Grid filter was applied to downsample the original vehicle point cloud data to obtain the diluted road point cloud data;Secondly,a direct filtering algorithm was explored to filter the thinned point cloud data,eliminating point cloud data such as tall buildings and vegetation,and a gradient filtering algorithm was used for separating ground and non ground points;Finally,the boundary feature estimation method was realized to extract the three-dimensional(3D)boundary points of the road.Experiments were conducted using two sets of point cloud data for different types of road sections.The results showed that the integrity and accuracy rates of the method in this paper for extracting road edges on straight road sections were 96.3％and 98.8％,while the integrity and accuracy rates for extracting road edges on curved road sections were 91.8％and 96.7％.The results showed that the method in this paper could effectively extract road boundary points,with high accuracy,and can provide reliable data support for high-precision map production.关键词
车载激光扫描/点云数据/边界点/边界特征估计法/高精地图Key words
vehicle mounted laser scanning/point cloud data/boundary points/boundary feature estimation method/high precision map分类
天文与地球科学
