期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2003
卷号：XXXIV-3/W13
出版社：Copernicus Publications
摘要：During the last decade laser scanning turned out to be one of the most promising techniques in photogrammetry. Some photogrammetrists even say that laser scanning has caused a paradigm change in photogrammetry because of its big impact on methods like DTM generation or 3D city model reconstruction. Originally, laser scanning was designed for DTM generation in forest areas. For that purpose several filtering approaches had been proposed and partly realized in commercial systems. This paper describes another alternative for filtering of laser scanning data. It combines a pre-filter based on a convex hull and a subsequent finite element adjustment. The pre-filter creates a crude TIN DTM assuming that the lowest measured laser points are representing the terrain accurately enough. The following adjustment of this TIN utilizes all laser points except non-ground points. This leads to a slight height change of the TIN points according to statistical and geometrical constraints and to a better approximation of the terrain. The method works hierarchically in a data pyramid starting from a coarse point density of about 4 m. The preliminary system has been successfully applied to laser scanning data that have been captured by the TopoSys system in a test site located in the National Park Bavarian Forest. The results show clearly that the method is successfully applicable in forests areas with different forest structures. The accuracy derived in controlled tests lies in the range of 15 – 25 cm and is also equivalent to the DTM quality of standard filtering procedures
