The analysis on the accuracy of DEM retrieval by the ground lidar point cloud data extraction methods in mountain forest areas

LiDAR data contains the elevation and brightness information of land surface, vegetation cover and construction. Ground filtering and interpolation method are the key for extracting the DEM accuracy based on the point clouds. This paper takes Zhangye City, Gansu Province in western mountainous areas as the study area, based on the point clouds of 0.7 points/m2, uses 5 m * 5 m grid screening method and the lowest Thiessen polygon point screening method to extract the ground point. Ordinary kriging interpolation method was used to retrieve Digital Elevation Model (DEM). Referring to the elevations of 1466 sample points, we analysed the accuracy for extracting DEM by the two selected methods of extracting ground point. The results showed that the DEM extracting accuracy by the near lowest point screening method is better than the grid screening method.