摘要
提出一种从机载LiDAR点云数据中自动提取建筑物平面的方法。给出了基于边长约束的三角形生长算法对建筑物初始区域进行提取,针对提取出的建筑物脚点,利用自适应MeanShift方法在特征空间中对其进行聚类分析,并提取出平面目标,最后利用Alpha-Shape算法生成建筑物平面的轮廓线。通过实验证实了方法的有效性。
A new method for extracting building planar objects from LiDAR data base on adaptive Mean Shift algorithm is proposed,which consists of three parts：building region detection from filtered LiDAR data using a triangle group algorithm constrained by triangle side length;planar object extraction from building foot prints based on cluster analysis in feature space by adaptive Mean Shift;planar object contour generation using Alpha-Shape algorithm.Experiment is conducted to demonstrate the efficiency of the proposed method.
出处
《计算机工程与应用》
CSCD
北大核心
2011年第10期5-7,共3页
Computer Engineering and Applications
基金
国家高技术研究发展计划(863)No.2006AA12Z101
No.2009AA12Z107~~
