面向对象的车载激光扫描点云建筑物立面提取

Object-based extraction of building facades from mobile LiDAR point clouds

本文利用面向对象的点云分析方法进行车载激光扫描点云数据中建筑物立面的识别。该方法对点云进行基于法向量和距离约束的点云分割,计算分割面片的倾角、面积、高程变化范围等特征,并依据建筑物立面垂直于水平面、面积比较大、高程变化范围较大等规则从点云数据中识别建筑物立面。实验表明该方法可以精确有效地进行MLS点云中建筑物立面提取。

An objected-based analysis method was proposed to extract building facades from the Mobile Laser Scanning（MLS）point clouds. First, point cloud segmentation was executed based on the two features about normal and distance constraint. Second, three characteristics of every segment about angle, area and elevation range were calculated. At last, the building facades were identified according to three predefined rules about perpendicular to the horizontal plane, a larger area and a larger elevation range. Experiments showed that the proposed method would be accurate and effective for building facades extraction from MLS point clouds.