点云模型分割及应用技术综述  

Review of Point Cloud Segmentation and Application

作　　者：刘进[1,2] 武仲科[1] 周明全[1] 

机构地区：[1]北京师范大学信息科学与技术学院,北京100875 [2]山东财政学院计算机信息工程学院,济南250014

出　　处：《计算机科学》2011年第4期21-24,71,共5页Computer Science

基　　金：国家"863"计划(2008AA01Z301);北京师范大学自主科研基金重点项目(2009SD-11)

摘　　要：介绍了点云模型分割的定义、分类和应用情况;比较、分析和评价了几类典型的点云模型分割算法,如边缘检测法、扫描线算法、聚类方法、基于图的分割方法等;给出了每种方法的应用特点和应用环境;指出了目前点云模型分割技术方法中存在的主要问题;分类介绍和评价了点云分割在相关应用中的研究工作;最后展望了点云分割技术今后的发展方向。A summary of the definition,classification and application of point cloud segmentation was presented,several kinds of typical point cloud segmentation algorithm were compared,analyzed and evaluated,such as edge-detection method,surface-growing method,scan-line method,clustering methods,graph partitioning methods,etc.The characteri-stics and application environments of each method were presented.The main problems in point cloud segmentation methods were pointed out,and the description and evaluation of point cloud segmentation in related fields were conducted.At last,problems and prospect of the techniques were also discussed.

关 键 词：点云 分割 激光扫描 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]