Laser point cloud filtering for DTM generation using kriging

This paper presents a method of filtering point clouds generated by laser scanning, to obtain a Digital Terrain Model (DTM). The filtering process is performed based on an approximated surface obtained from urban road points. These points are sampled in straight lines detected by Steger in the intensity image of the laser pulse. The main assumption of the method is that the ground has smooth behavior inside the block, so the sample laser points collected along the urban roads allow, using the kriging interpolation method, a suitable representation of the land inside the block, that is, relatively close to the ground point laser in these regions. Thus, filtering is performed by proximity of the original cloud laser points with approximate surface. For thus a DTM is obtained from the new sample by kriging interpolation method, increasing the description of the surface. From the experiments it was possible to verify the feasibility of the proposed method, with results of good visual consistency and satisfactory numerical indicators.