Research On Building Information Extraction Algorithm Based On Airborne LiDAR Point Cloud Data

Airborne LiDAR Technology(LiDAR:Light Detection and Ranging)has developed rapidly in recent years and has been widely used in 3D City modeling.The fast,accurate and efficient acquisition of 3D point cloud information by airborne LiDAR technology provides the data base for 3D urban modeling.How to extract the building information from the 3D point cloud data is a key problem.Based on the research of LiDAR point cloud data of elevation characteristics and geometric distribution characteristics,summarizes the existing point cloud filtering algorithm and building information extraction method,combined with the characteristics of point cloud data,improved TIN network filtering algorithm and plane fitting method,the extraction of high precision can achieve the non ground point accurate separation and building information.The main research contents include the following aspects:(1)Improved TIN(triangulated irregular network)filtering algorithm.According to the elevation difference between the different points of the cloud,the different height difference threshold is set up,and the separation between the ground points and the object points is combined with the TIN filter algorithm.The method can accurately separate the point of the surface and the surface of the object,and effectively preserve the target point cloud data of buildings and other objects.(2)Building point cloud extraction based on slope change characteristics of building roof point cloud data.Most of the roof of a building is a regular geometric figure,and it has an approximate horizontal,inclined plane or a smooth surface.Therefore,by controlling the range of the slope and the horizontal distance of the adjacent point cloud,we can extract the point cloud of the approximate horizontal plane,inclined plane or continuous smooth surface.This method can extract the information of the building roof.(3)Alpha-shape algorithm is used to extract the edge of the building,and the contour line is fitted.Select the part of the roof of the building shape of point cloud detected by-shape method and edge points and edge points will be carried out by building contour line of least squares fitting,finally through the rule building contour contour approximation.The experimental results show that the method can accurately extract the contour of the building roof,and approach the real building contour.The Terrasolid software is selected to test the algorithm,and the results are compared with the results of Terrasolid software.The experimental results show that the proposed algorithm can effectively preserve the non ground point cloud information,and accurately extract the roof of the building point cloud.