摘要
利用机载LiDAR等三维激光扫描设备对城市区域进行范围扫描,获取的点云数据可通过点云分类的方式来提取建筑的几何要素,进而应用于城市测绘领域。针对人工交互分类方式效率低下等问题,实现了一种主要面向建筑屋面和外墙要素的点云自动分类算法。首先根据训练区域的点云局部特征构建特定的特征标签,再采用分类识别算子结合点云局部非平面性、高程、垂直度、RGB值等特征要素,完成建筑屋面和外墙的分类。实验测试结果表明,算法能较明确地完成建筑屋面和外墙点云的分类,且分类效率远高于人工交互分类方式。
Using airborne LiDAR and other 3D laser scanning equipment to scan the urban area,the obtained point cloud data can be used to extract the geometric elements of buildings by the way of point cloud classification,and then applied to the urban surveying and mapping.Aiming at the low efficiency of manual interactive classification method,we put forward an automatic point cloud classification algorithm for building roof and exterior wall elements.We constructed a specific feature label according to the local features of point cloud in the training area at first.Then,combining with the local non planarity,elevation,verticality,RGB value and other feature elements of point cloud,we used a variety of classification and recognition operators to complete the classification of building roof and exterior wall.The experimental results show that the classification result of this algorithm is clear,and the classification efficiency is much higher than that of manual interactive classification method.
作者
贾军辉
杨涛
沈大勇
付昕乐
JIA Junhui;YANG Tao;SHEN Dayong;FU Xinle(The Third Institute of Photogrammetry and Remote Sensing,Ministry of Natural Resources,Chengdu 610100,China)
出处
《地理空间信息》
2022年第9期67-71,共5页
Geospatial Information
基金
四川省科技厅重点研发资助项目(17ZDYF1099)。
关键词
城市点云
建筑屋面
外墙
点云分类
urban point cloud
building roof
exterior wall
point cloud classification
