结合OTSU与迭代三角网的机载LiDAR建筑物点云提取  

Aerial LiDAR Building Point-cloud Extraction Algorithm Combining OTSU and Iterative TIN

作　　者：洪绍轩 王竞雪[1,2] HONG Shaoxuan;WANG Jingxue(School of Geomatics,Liaoning Technology University,Fuxin,Liaoning123000,China;Faculty of Geosciences and Environmental Engineering,Southwest Jiaotong University,Chengdu610031,China)

机构地区：[1]辽宁工程技术大学测绘与地理科学学院,辽宁阜新123000 [2]西南交通大学地球科学与环境工程学院,成都610031

出　　处：《遥感信息》2018年第6期79-85,共7页Remote Sensing Information

基　　金：国家自然科学基金(41871379);辽宁省教育厅科学研究一般项目(LJYL010)

摘　　要：针对机载LiDAR建筑物点云提取过程中受地形影响参数设置困难,建筑物、树木难以区分等问题,提出一种结合最大类间方差法与迭代三角网相结合的机载LiDAR建筑物点云提取算法。在已有滤波结果的基础上,首先采用最大类间方差法对滤波得到的非地面点进行预处理,提取初始建筑物点;然后运用改进的迭代三角网方法对初始建筑物点云进行精确提取,得到最终的建筑物点云。实验选取国际摄影测量与遥感协会提供的三组LiDAR点云数据进行建筑物点云提取。结果表明,该算法可以较好地实现建筑物点云的高精度自动提取,且对不同屋顶类型以及地形具有良好的自适应性,验证了算法的可靠性。Aiming at the problem that setting parameter difficultly due to the influence of topography,and it is difficult to distinguish between buildings and trees,a building point-cloud extraction algorithm combining OTSU and iterative TIN for aerial LiDAR point-clouds is proposed.On the basis of filtering,the algorithm makes pretreatment for the non ground points after filtering by the OTSU firstly to get initial building point-clouds.Then,extract building point-clouds precisely for the initial building point-clouds by the improved iterative TIN algorithm.Finally,obtain buildings point-clouds.The experiment selected three sets of LiDAR point-cloud dataset provided by the International Society for Photogrammetry and Remote Sensing to extract building point-clouds,and the results indicated that the proposed algorithm can achieve building point-clouds extraction accurately and automatically;it also has fine adaptability for different roof types and topography,verifying the reliability of the building point-cloud extraction algorithm.

关 键 词：机载LIDAR 滤波 建筑物点云提取 最大类间方差法 迭代不规则三角网 

分 类 号：P237[天文地球—摄影测量与遥感]