LiDAR点云数据分割的自适应回波比率算法研究  

Slope adaptive echo ratio algorithm for LiDAR segmentation in tilt roof

作　　者：杨晓云 岑敏仪[1] 梁郁 梁鑫[1,2] YANG Xiaoyun;CEN Minyi;LIANG Yu;LIANG Xin(Faculty of Geosciences and Environmental Engineering,Southwest Jiaotong University,Chengdu 610031,China;Faculty of Civil Engineering,Guangxi University of Science and Technology,Liuzhou,Guangxi 545006,China)

机构地区：[1]西南交通大学地球科学与环境工程学院,成都610031 [2]广西科技大学土木建筑工程学院,广西柳州545006

出　　处：《测绘科学》2018年第7期128-134,148,共8页Science of Surveying and Mapping

基　　金：广西高等教育教学改革工程项目(2014JGA188);广西高校科学技术研究项目(YB2014213);广西科技大学科学科研项目(校科自1307103)

摘　　要：回波比率描述了地物表面的激光点云在垂直方向分布的离散程度和穿透力,可用于分割建筑物和植被脚点。针对传统的回波比率计算方法受实体表面坡度的影响较大,通常随着地物表面倾角的增大,其计算值呈下降趋势,该文在B.Hfle自适应坡度回波比率算法基础上,通过对邻域体系分析发现3D与2D邻域的垂直方向参数尺度不一致才是造成倾斜表面回波比率减小的真实原因。采用不同的空间邻域进行回波比率计算,并在此基础上提出了改进的自适应坡度回波比率计算方法。较B.Hfle算法,改进后的方法无须计算局部表面坡度,仅调整3D邻域的垂直方向参数尺度就可适用于不同坡度倾斜屋顶表面。通过实测LiDAR点云进行试验验证,新算法的正确率和效率较原有算法都有较大的提高。Echo ratio describes the dispersion of LiDAR distribution and object transparency,so it is always used as significant parameter for distinguishing buildings from trees.The present computation of ER is affected by surface slope.In general,ER value reduces in trend with the increasing of obliquity.By analysis of neighborhood for ER computation,vertical-parameter inconsistency of 3D and 2D cause ER decreasing for the tilt surface.This paper deduces an improved method with different neighborhood.For applying to variety tilt roof,the new method adjusts vertical parameter in 3D neighborhood,without computing local surface slope.Measured LiDAR data is used to test the accuracy and efficiency of the new method.The segmentation result improves greatly comparing with B.Hfle method.

关 键 词：机载激光雷达 邻域体系 自适应坡度 回波比率 分割 

分 类 号：P208[天文地球—地图制图学与地理信息工程]