River point cloud data down-sampling method based on density layering

Abstract

The invention discloses a river point cloud data down-sampling method based on density stratification. The method comprises the following steps: step 1, obtaining point cloud data of a river scene; 2, sparse point cloud data of the river scene are identified and stored; 3, dense point cloud data of the river scene are identified and layered; and step 4, performing point cloud down-sampling of different degrees on different layers in the dense point cloud data: performing point cloud down-sampling of different degrees on different layers in the dense point cloud data based on the dense point cloud data subjected to layering processing in the step 3. Therefore, the downsampling of the river point cloud data is controllable, the sparsity of the ship point cloud can be prevented from being aggravated due to the downsampling of the data, the invalid river bank point cloud with redundant points can be effectively removed, the difference between the foreground points and the background points in the river point cloud is reduced, and the accuracy of the river point cloud is improved. And ship target detection in the scene is facilitated.

Images