车载LiDAR点云中建筑物立面的自动分割  

An automatic segmentation method for building facades from vehicle-borne LiDAR point clouds

作　　者：李永强[1,2] 吕海洋[3] 苏蕾[1] 王文越[1] 曹鸿[1] 杨莎莎[1] 

机构地区：[1]河南理工大学测绘与国土信息工程学院,河南焦作454000 [2]北京师范大学减灾与应急管理研究院,北京100875 [3]南京师范大学虚拟地理环境教育部重点实验室,南京210046

出　　处：《测绘科学》2014年第4期106-109,共4页Science of Surveying and Mapping

基　　金：国家自然科学基金(41001304);科技部973课题(2011CB707102);国家十二五科技支撑计划项目(2012BAH34B);河南理工大学博士基金(B2009-33)

摘　　要：将建筑物立面数据从散乱无序的车载LiDAR点云中分割出来是一项繁琐而艰巨的任务。本文提出基于基础地理数据辅助的分割方法:首先对基础地理数据进行要素精简,提取有效的建筑物轮廓数据;然后将提取数据转换到车载LiDAR点云数据所在坐标系中,实现坐标基准的统一;再根据建筑物立面特征设置合理的建筑物轮廓缓冲阈值,对建筑物立面点云进行自动分割;最后采用合理的质量评价机制,对分割结果进行检核和评价,实现分割结果的质量控制。实验表明文中方法简单有效。It＇s a tedious and difficult task to segment the building facades data from vehicle-borne Li-DAR point cloud data, which is commonly considered as scattered and disorderly point clouds. In this paper, the author proposed a segmentation method based on the fundamental geographic data： firstly, simplify the features of fundamental geographic data, and extract effective contour information of the buildings; secondly, convert the coordinate system of fundamental geographic data to that of Vehicle-borne Li- DAR point cloud to realize the unification of coordinate system; thirdly, set a suitable buffer threshold value for building contour, and segment out the point cloud data of building facades automatically; lastly, take a reasonable quality assessment mechanism to check and evaluate the segmentation results with their quality control. Experiment shows that the proposed method would be simple and effective.

关 键 词：车载LiDAR 点云 基础地理数据 建筑物立面 自动分割 

分 类 号：P225.2[天文地球—大地测量学与测量工程] TP75[天文地球—测绘科学与技术]