基于体素的森林地区机载LiDAR数据DTM提取  

A voxel-based filtering algorithm for DTM data extraction in forest areas

作　　者：唐菲菲[1,2] 刘经南[3] 张小红[1] 阮志敏[4] 

机构地区：[1]武汉大学测绘学院 [2]重庆大学土木工程学院 [3]武汉大学卫星导航定位技术研究中心 [4]重庆交通科研设计院勘察设计所

出　　处：《北京林业大学学报》2009年第1期55-59,共5页Journal of Beijing Forestry University

基　　金：国家自然科学基金项目(40504001);遥感国家重点实验室资助项目"利用机载激光雷达数据提取森林地区DTM"

摘　　要：机载LiDAR是一种能够直接、快速获取被测目标三维空间信息的主动式遥感技术,被广泛用于获取高精度的数字地面模型,但是在植被比较密集、地势比较陡峭的森林地区,能够穿透植被到达地表的激光脚点数量较开阔区域少,对于提取精确的DTM有一定难度。该文提出一种继承式多分辨率体素滤波算法,从机载激光扫描数据中获取森林地区的DTM。该方法将激光点云数据划分为不同分辨率等级的体素,以体素为单位通过与邻域体素的高程加权均值比较,剔除植被点,保留地面点,从而获取森林地区的DTM。实验证明该滤波方法能够有效地提取森林地区的DTM。LiDAR is an advanced remote sensing technique, which can collect the 3D spatial information of objects directly and effectively, and it has been widely used in acquiring precision digital terrain model （ DTM）. Nevertheless, in forest areas with a dense vegetation and steep slopes, the amount of laser footprints that can penetrate through the vegetation and reach the ground is relatively less than that in an open area, so the DTM acquisition is difficult to some extent in forest areas. The paper proposes an inherited multi-resolution voxel-based filtering algorithm, aiming at deriving the digital terrain model in forest areas. The laser scanning data were divided into voxels with different resolutions by comparing the weighted average height with neighbors. Vegetation in voxels was rejected and terrain points were retained to be interpolated into the DTM. This study suggests that the filtering algorithm is effective in DTM acquisition of forest areas.

关 键 词：机载激光扫描 继承式多分辨率 体素 滤波 数字高程模型 

分 类 号：S757.2[农业科学—森林经理学]