基于无人机激光雷达的建筑变化立体检测OA北大核心CSTPCD

In the detection of building changes in three dimensions,considering the complex shape of buildings,the large volume and uneven distribution of point cloud data,a significant noise issue is encountered in the acquired building point cloud images.By applying drone-based LiDAR technology,a novel method for detecting building changes in three dimensions is designed.The point cloud data of buildings are collected through drone-based LiDAR technology,and the concept of multi-scale is introduced.A multi-level virtual grid is established to determine the grid spacing and grid specifications.A bidirectional filtering algorithm is employed to perform filtering processing layer by layer according to the grid scale size,achieving data preprocessing.Through registration processing,matching point pairs are obtained from the reconstructed data,and the three-dimensional spatial distance of corre-sponding points is calculated.By using distance thresholds and judgment criteria,it is determined whether the point cloud data has changed,thereby realizing the detection of building changes in three dimensions.Experimental results demonstrate that the pro-posed method can accurately and comprehensively detect building changes,with an accuracy rate as high as 90％,and with mini-mal detection time.