基于扫描线坡度统计的LiDAR滤波算法  

LiDAR Filtering Based on Slope Statistic of Scanning Line

作　　者：杨晓云[1,2] 梁鑫[1] 

机构地区：[1]广西工学院土木建筑工程系,广西柳州545006 [2]西南交通大学土木工程学院,四川成都610031

出　　处：《广西工学院学报》2011年第4期65-68,共4页Journal of Guangxi University of Technology

基　　金：广西教育厅科研项目(200911LX161);广西工学院科学科研项目(院科自0840209)资助

摘　　要：提出一种基于扫描线坡度统计的LiDAR滤波算法,该方法根据扫描线内点间距较小、密度较大的特点,将二维滤波转化为一维滤波.在统计扫描线内相邻点间的坡度,设定阈值迭代滤除地物点基础上,采用局部参数化表面拟合进一步将精化候选地面点,用ISPRS提供的参考数据进行实施测试分析,结果表明:可有效处理大面积、复杂城区,为下一步城市的三维重建打下很好的基础.This paper proposes a new filter based on slope statistic to extract the ground points from airborne LiDAR data. For the high point density in scanning line, the method transferred two-dimensional filtering into one-dimensional, which improved the algorithm＇s efficiency greatly. Firstly, slope between two points in scanning line had been worked out. By slope statistic, non-ground points were removed. Secondly, curved surface fitting was used to correct the wrong detection of ground points. Reference data sets provided by ISPRS were selected to test the filter based on slope statistic. The data came from city areas. The results showed that the filter can remove most of the non-ground points effectively.

关 键 词：LIDAR 扫描线 坡度统计 滤波 

分 类 号：P237.3[天文地球—摄影测量与遥感]