Abstract: In response to the incomplete evaluation elements of existing point cloud accuracy and the inability to effectively control the accuracy of point cloud data processing, a comprehensive evaluation method for point cloud acquisition and data processing accuracy based on point, line, and surface elements is proposed for the first time to meet the application requirements of point cloud data in water and drought disasters. The point cloud plane dimension proposes a comprehensive evaluation of the point cloud plane accuracy based on factors such as point plane accuracy, feature lines of land features, single point plane position of land features, edge length and area of land features.The elevation dimension of point clouds is proposed to comprehensively evaluate the accuracy of point cloud elevation by considering factors such as single point elevation, terrain profile, and terrain surface model, including the enclosing area of terrain profiles and terrain volume. The results indicate that the laser point cloud comprehensive accuracy evaluation method based on water and drought disasters can comprehensively, objectively, and accurately evaluate the accuracy of point clouds from the plane and elevation dimensions, as well as the entire process of point cloud data acquisition, point cloud filtering, and data application. It can be extended to other fields of laser point cloud applications and has good reference significance.

Key words: water disasters, laser point cloud, comprehensive accuracy, accuracy evaluation