基于点云特征的建筑物屋顶面提取

Extraction of Building Roof Surface Based on Point Cloud Features

针对从激光雷达点云中提取建筑物屋顶面精度不高、效果不佳等问题,提出了一种利用点云特征的建筑物屋顶面提取方法。通过统计点云的屋顶面法向量特征识别屋顶面,利用改进的基于密度的空间聚类算法实现屋顶面聚类,并采用张量投票的方式解决屋顶面点云竞争的问题。实验结果表明,该方法能有效提取屋顶面点云,总体精度约为96%,能满足实际工作的需要。

Aiming at the problem of low accuracy and poor effect of extracting building roof surface from LiDAR point cloud,we proposed a building roof surface extraction method based on point cloud features.We identified the roof surface by collecting the point cloud normal vector features of roof surface,used the improved density-based spatial clustering of applications with noise(DBSCAN)algorithm to realize the roof surface clustering,and used the tensor voting method to solve the problem of roof surface point cloud competition.The experimental results show that the method can effectively extract the point cloud of roof surface,and the overall accuracy is about 96%,which can meet the needs of actual work.