METHOD FOR ENHANCING THREE-DIMENSIONAL SONAR POINT CLOUD IMAGE BASED ON DISTANCE STATISTICS

摘要

Provided is a method for enhancing a three-dimensional sonar point cloud image based on distance statistics, comprising the following steps: (1) acquiring sonar data, and converting three-dimensional sonar range image information corresponding to each frame of the sonar data into point cloud data in a global coordinate (S01); (2) establishing a kd-tree for N pieces of point cloud data, and searching for each piece of point cloud data Pi by using the kd-tree (S02); (3) utilizing a K-nearest neighbour algorithm to search K pieces of nearest point cloud data of the point cloud data Pi, and calculating a Euclidean distance Lij between the point cloud data Pi and the K pieces of point cloud data thereof (S03); (4) calculating an average value Li of K Euclidean distances Lij belonging to the point cloud data Pi (S04); (5) calculating a mean value μ and a standard deviation σ of N Li values (S05); and (6) carrying out statistics on all the Li values by utilizing Gaussian distribution with the mean value being μ and the standard deviation being σ, and removing point cloud data corresponding to the Li falling out of the range a-b (S06). The method is easy to operate, and is efficient and quick, and can effectively remove outliers, thereby achieving the purpose of reducing noise and enhancing a point cloud image.