智能城市2026，Vol.12Issue(4)：79-83,5.DOI:10.19301/j.cnki.zncs.2026.04.018
基于三维激光点云的规划竣工建筑立面自动提取研究
Research on automatic extraction of planned and completed building facades based on 3D laser point cloud
杨坤1
作者信息
- 1. 北京富地时空信息技术有限公司,北京 100013
- 折叠
摘要
Abstract
To address the problem of inaccurate extraction of building facades in planning completion caused by shadows and distortion in oblique photographic images,the article proposes an automatic extraction method for planning completion building facades based on 3D laser point clouds.The method uses 3D laser scanning technology to obtain point cloud data of the planning completion buildings and performs 3D point cloud segmentation based on the best-fitting plane.The point clouds of the segmented building planes are meshed,and the distance between the point clouds and the mesh center as well as the point cloud elevation are calculated to extract building point cloud features.These point cloud features are input into a support vector machine classifier using a linear kernel function to achieve automatic extraction of building facades.The results show that the deviation distance of the facade extraction using this method is less than 0.03 m,realizing precise extraction of building facades.关键词
三维激光点云/规划竣工建筑/立面提取/点云分割/边界特征/支持向量机Key words
3D laser point cloud/planning and completing buildings/facade extraction/point cloud segmentation/boundary features/support vector machine分类
信息技术与安全科学引用本文复制引用
杨坤..基于三维激光点云的规划竣工建筑立面自动提取研究[J].智能城市,2026,12(4):79-83,5.
