改进的参数统计机载雷达点云滤波方法  

An improved parametric statistics filtering for airborne LiDAR point clouds

作　　者：胡永杰[1] 程朋根[1,2] 陈晓勇[1] 李荣[1,2] 辛俊伟[1] 

机构地区：[1]东华理工大学测绘工程学院,南昌330013 [2]广西空间信息与测绘重点实验室,桂林541004

出　　处：《测绘科学》2015年第11期39-42,共4页Science of Surveying and Mapping

基　　金：国家自然科学基金资助项目(41161069);广西空间信息与测绘重点实验室资助课题(13-051-14-18;11-031-08-26);东华理工大学校长基金项目(DHXK201401)

摘　　要：针对基于高程的偏度参数统计的LiDAR数据滤波对地形起伏城区滤波适应性不强的弱点,文章对该方法进行了改进,提出基于二次曲面高程残差的偏度参数统计法的LiDAR数据滤波。实验结果表明改进的方法不仅延续了参数统计法无需阈值的优势,而且对城区滤波更具适应性、合理性和稳定性,能够明显提高滤波精度。Aiming at the problem that the adaptability of the filtering algorithm based on the statistical skewness parameters of elevations for LiDAR data is not strong for urban topography,the paper proposed an improved method of skewness parametric statistical filtering based on quadric elevation residuals.Experimental result showed that the method could have not only the advantage of the parametric statistical methods which needs no threshold,but also more adaptability,rationality and stability with higher precision of filtering.

关 键 词：LiDAR滤波 格网化 二次曲面 高程拟合残差 偏度系数 

分 类 号：P228[天文地球—大地测量学与测量工程]