一种车载激光扫描数据的滤波处理算法

A Filtering Method of Vehicle-Borne Laser Scanning Data

车载激光扫描系统能够快速、高效地获取城市地物的三维空间信息,是采集城市近景三维空间信息的一个重要手段。设计了一种从车载激光点云中分离出地面点和非地面点的滤波算法。首先建立规则格网,基于格网中激光点的高程差提取出初始地面点的方法;然后基于格网分层过滤出精确的地面点,从而达到地面点和非地面点的分离。实验结果表明:该算法滤波效果良好,能够满足城市DTM建立的精度要求。

Vehicle-borne laser scanning system can obtain three-dimensional（3D） information of urban objects quickly and efficiently. It is an important method of acquiring the close-range 3D data of cities. A kind of filtering algorithm was designed in this paper, which could separate out the ground points and non-ground points from the vehicle laser points cloud. Firstly, the regular grid is established, and then the initial ground points are extracted based on the elevation difference of the laser points in the grid. Secondly, the accurate ground points are filtered based on the layered grid, so as to separate the ground points and non-ground points. The experimental results show that the proposed method can get good filtering effect and satisfy the accuracy requirements of establishing the city DTM.