Simulation of Point Cloud Information Collection Method for Building Panoramic 3 D Airborne Radar
Due to measurement errors of airborne radar and environmental interference,the collected point cloud data may contain noise and missing values.In addition,the multi-layer structure of buildings may cause position devi-ation or attitude differences between point cloud data,resulting in large errors in the collection of panoramic point cloud information.To address this issue,this article designed a simulation of collecting point cloud information of pan-oramic 3D airborne radar of buildings.Firstly,we adopted the bidirectional cloth simulation method to extract the point cloud of the building's top surface.Then,we used the normalized digital surface models penetrability analysis,morphological opening operation,and constrained growth method to extract the panoramic point cloud of buildings.Next,we used statistical filtering algorithms to filter outlier noise.Meanwhile,we used a cloud simplification algorithm including voxel centroid adjacent feature points to reduce the redundancy of the point cloud and perform rough and fine registration of the panoramic point cloud.Finally,we achieved the collection of panoramic 3D point cloud infor-mation.Experimental results show that the proposed method has smaller errors,higher completeness rates,and higher collection quality.
3D architectural panoramic viewAirborne radarPoint cloud structure information collectionBi-directional fabricICP algorithm
NETL
NSTL
万方数据
