Abstract: Aiming at the advantages and disadvantages of LiDAR point cloud and UAV image data, the LiDAR point cloud and DOM image of UAV are combined to transfer the spectral information of image data to LiDAR point cloud data, so that it not only has accurate spatial structure information, but also can obtain clear texture information. In order to verify the feasibility of fusion data application and the accuracy of data extraction, ground point extraction and DEM construction are carried out on point cloud data before and after fusion. The experiment shows that the LiDAR point cloud data can be extended from four dimensions to seven dimensions by transferring the spectral information of UAV image to the LiDAR point cloud data. After the fusion, the point cloud data has clear texture information, the ground object type is easier to interpret, and the ground points are separated completely. Through the comparative analysis of DEM model, the expression of DEM model constructed by integrating point cloud data is closer to the real surface. The research results provide some technical support for the further application of multi-source point cloud data.

Key words: airborne LiDAR, point cloud, UAV images, data fusion, model construction