 Object-based point cloud analysis, OBPA, is a powerful tool for extracting information from airborne LiDAR point clouds. It uses a support vector machine, SVM, to classify the point clouds into different categories based on their geometry, radiometric, topological, and echo characteristics. The method was tested on three datasets with varying levels of complexity and achieved high accuracy rates of over 92%. This suggests that OBPA can be used to accurately classify urban point clouds, regardless of their density. This article was authored by Xiao Jiangming, Xiang Wulin, and Ji Xinjiang.