基于密度噪声应用空间聚类算法的机载激光雷达建筑物点云提取与单体化

针对机载激光雷达建筑物点云提取过程中自动化提取困难,以及提取后的建筑物单体化过程烦琐等问题,提出一种基于密度噪声应用空间聚类(density-based spatial clustering of applications with noise,DBSCAN)算法的机载雷达建筑物点云提取与单体化的方法.该方法对预处理后...

Point Cloud Extraction and Monomer of Airborne LIDAR Buildings Based on DBSCAN Algorithm

Aiming at the difficulty of automatic extraction in the process of airborne lidar building point cloud extraction and the tedious process of building monomer extraction, a method of airborne lidar building point cloud extraction and monomer based on DBSCAN algorithm is proposed. In this method, the preprocessed point cloud data is denoised and preliminarily extracted based on DBSCAN algorithm. Through 3D density clustering, the point cloud of buildings is extracted and automatically individualized. According to the characteristics of building point cloud density, two-dimensional density clustering is carried out, and the point cloud is segmented with dom. Finally, the processed point cloud data is optimized, and the single building cluster class is extracted to get the single building point cloud. The results show that the accuracy rate of the number of building point clouds extracted by the extraction method is 97.36%, and the mean square error of the contour side length is 0.077, which can effectively extract the building point clouds and make them monomer.