由机载激光扫描系统获取的点云数据,可以直接生成扫描区域的数字表面模型,但要想提取数字地面模型还须对点云数据进行分类处理。提出了一种获取数字地面模型的方法,其主要思路是利用原始的激光数据构建二维三角网,从而构建数字表面模型。通过分析数字表面模型的坡度进行初始的分割,剔出坡度较大的三角形。经过初步的分割之后,通过连通区域分析来获取每个分割区域的特征,提取出裸露地表高程,从而由点云数据构建出数字地面模型。

Airborne LIDAR can collect the point cloud data and construct the surface model of the objects on the earth.But for the production of digital elevation model(DEM) from the point cloud,the classification(filtering),feature extraction and quality control still pose great challenges.For extraction of DEM in urban area,the segmentation of the point cloud should be carried out.In this article,a method for deriving the DTM based on slope analysis is presented.Firstly,a TIN model is constructed from the original point cloud.The slopes are calculated in the TIN model.Then the segmentation is carried out based on the slopes in the TIN.The triangular region with large slope value is deleted.After the segmentation,the region-growing processing is used to merge the small regions according to the connectivity.The part which can cover the biggest area is considered to be the road.Finally,the terrain model is generated by kriging interpolation with these points.