机载LiDAR点云建筑物边界线规则化算法研究

Research on buildings boundary regularization algorithm for aerial LiDAR point clouds

针对单一参数的Alpha-Shape算法无法适应密度差异较大的点集数据以及提取的边界信息具有锯齿形状的问题,结合Alpha-Shape算法与D-P算法进行轮廓线的粗略提取,利用最小二乘方法以及建筑物边界线之间向量、长度等特征确定建筑物关键点;通过寻找建筑物主方向,实现建筑物边界线规则化。实验选取国际摄影测量与遥感协会提供的典型区域的LiDAR点云数据进行建筑物边界线提取,并与传统Alpha-Shape算法提取建筑边界线结果进行比较,结果表明本文算法在建筑物边界线信息提取方面更准确、更稳定。

In order to solve the problem that the single parameter Alpha-Shape algorithm cannot adapt to the point set data with large density difference and the saw tooth shape of boundary information.Firstly,the Alpha-Shape algorithm and D-P algorithm are combined to extract the contours roughly.Then the least square method and the characteristics of vector and length between the building boundary lines are used to determine the key inflection points of the building.Finally,by finding the main direction of the building,the regulation of the boundary line of the building is realized.Typical LiDAR point clouds data,which are provided by the International Society for Photogrammetry and Remote Sensing,are selected for buildings boundary regularization extraction experiments.The results are compared with those of the traditional Alpha-Shape algorithm in extracting the building boundary line.The results show that the proposed algorithm is more accurate and stable in building boundary information extraction.