车载激光点云组合滤波算法
Combined filtering algorithm for vehicle-mounted laser point cloud
白建霞 1姚登峰2
作者信息
- 1. 浙江省国土勘测规划有限公司,浙江 杭州 310012
- 2. 浙江省测绘科学技术研究院,浙江 杭州 311100
- 折叠
摘要
为了解决传统车载点云滤波算法无法有效剔除地形坡度较大区域非地面点的问题,本文结合布料模拟滤波(CSF)算法与传统不规则三角网(TIN)的特点,提出一种基于CSF算法的改进TIN滤波算法.首先,使用CSF算法对原始点云进行滤波处理,获取初始地面点;其次,使用改进TIN滤波算法对CSF算法滤波结果进行进一步滤波,通过邻域卷积和选取种子点与构造虚拟种子点,根据滤波次数改变格网大小进行迭代滤波.使用两组地形条件不同的城市道路点云进行试验,对比不同滤波算法的滤波结果.结果表明,本文提出的组合滤波算法能够有效弥补单一滤波算法的不足,能够较好地适应城区情况.
Abstract
Traditional vehicle-mounted point cloud filtering algorithms are unable to effectively remove non-ground points in areas with large terrain slopes.To solve this problem,this paper analyzed the characteristics of the cloth simulation filtering(CSF)algorithm and the traditional triangular irregular network(TIN)and proposed an improved TIN filtering algorithm based on the CSF algorithm.Firstly,the CSF algorithm was used to filter the original point cloud,and initial ground points were obtained.Secondly,the improved TIN filtering algorithm was used to further filter the results filtered by the CSF algorithm.Neighborhood convolution was utilized to select seed points and construct virtual seed points,and the grid size was changed according to the number of filtering times for iterative filtering.Two sets of urban road point clouds with different terrain conditions were used for experiments,and the filtering results of different filtering algorithms were compared.The results show that the combined filtering algorithm proposed in this paper can effectively compensate for the shortcomings of a single filtering algorithm and can better adapt to urban conditions.
关键词
点云滤波/布料模拟滤波/改进不规则三角网/格网化/精度检验Key words
point cloud filtering/cloth simulation filtering/improved irregular triangular networks/grid/accuracy inspection引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据