利用RANSAC算法对建筑物立面进行点云分割Segmentation of building facade point clouds using RANSAC
李娜;马一薇;杨洋;高晟丽;
摘要(Abstract)：
建筑物立面点云分割是车载激光扫描数据特征提取与建模的基础。本文将随机抽样一致性算法(Ran-dom Sampling Consensus)方法引入对点云的分割中,并在判断准则中引入了点云的r半径密度,消除了噪声的影响,同时建立角度和距离两个约束条件对平面分割结果进行优化,提取出了最终的建筑物立面特征平面。
关键词(KeyWords)： 车载激光扫描;随机抽样一致性;点云分割;r半径密度
基金项目(Foundation):
作者(Authors): 李娜;马一薇;杨洋;高晟丽;
DOI: 10.16251/j.cnki.1009-2307.2011.05.017
参考文献(References)：
- [1]李清泉.三维空间数据的实时获取,建模和可视化[M].武汉:武汉大学出版社,2003.
- [2]史文中,李必军,李清泉.基于投影点密度的车载激光扫描距离图像分割方法[J].测绘学报,2002,32(2):95-100.
- [3]Wani M A,Arabnia H R.Parallel edge re gion-basedsegmentation algorithm targeted at reconfi-gurable multi-ring network[J].Journal of Supercomputing.2003,25(1):43-62.
- [4]Sithole G,Vosselman G.Automatic structure detection ina point-cloud of an urban landscape[C]//2nd JointWorksho Pon Remote Sensing and Data Fusion over Ur-ban Aereas(Urban 2003).
- [5]Rabbani T,Heuvel F A,Vosselman.SEGMEN-TATIONOF PO-INT CLOUDS USING SMOOTH-NESS CONST-RAI-NT[J].*in IAPRS Volume XXXVI,Part5.Dresden.2006.
- [6]Fischler M A,Bolles R C.Random Sam-ple Consensus:A Paradigm for Model fitting with ap-plication to ImageAnalysis and Automated Cartogra-phy[J].Communica-tions of the ACM,1981,24(6):381-395.
- [7]陈付幸,王润生.基于预检验的快速随机抽样一致性算法[J].软件学报,2005,16(8):1431-1438.
- [8]Bauer J,Karner k,Schindler K,Klaus A,ZachC.Segmentation of building from de-nse 3D point-clouds.in Proceedings of the ISPRS[M].WorkshoPLaser scanning.2005.Enschede,the Netherlands.