基于PCA法矢估计的建筑点云立面边界提取

Building Point Cloud Elevation Boundary Extraction Based on PCA Normal Vector Estimation

针对建筑物点云在工程应用中存在大量数据不产生作用,只需保留建筑边界点云的问题,本文提出一种基于主成分分析法向量估计的建筑点云立面边界提取方法。该方法首先采样统计离群值去除算法对点云数据进行滤波去噪,然后采用主成分分析法估计样点表面的法线方向,最后使用角度阈值确定建筑物点云立面的边界。通过实例分析,此法能够较完整地提取建筑点云的边界特征,为相关工程应用提供了技术支撑。

For the point cloud of buildings,there is a large amount of data in engineering applications that has no effect,and only the point cloud of the building boundary needs to be retained.This paper proposes a method for extracting the boundary of the building point cloud elevation based on principal component analysis normal vector estimation.In this method,the sampling and statistical outlier removal algorithm is used to filter and denoise the point cloud data,and then the PCA method is used to estimate the normal direction of the sample point surface.Finally,the angle threshold is used to determine the boundary of the point cloud facade of the building.Through case analysis,the boundary features of the architectural point cloud can be extracted more completely,which provides technical support for related engineering applications.