Point cloud feature extraction method based on feature histogram in point cloud matching

Abstract

The invention discloses a point cloud feature extraction method based on a feature histogram in point cloud matching. In the point cloud matching, firstly, feature points are extracted, then, the feature points are described, and finally, matching is performed through a descriptor. In the invention, a vertical line of the surface taking the point as the center is utilized; the concavity and convexity of the local curved surface are solved by combining the surface with the normal of the point, so that the interference of some noise points is reduced. Secondly, when the triangular surface around the central point is smaller than 2, the point is an edge point, and interference of the edge point on matching is removed to a certain extent; in the description of the feature points, the depth, distance and local bending degree of the points are integrated, a neighborhood point set is subjected to gradient processing, and the gradient of the points is solved, so that the purpose of feature description of the points is achieved.

Description

technical field [0001] The invention belongs to the technical field of point cloud matching, and the invention relates to a point cloud feature extraction method, in particular to a feature histogram-based point cloud feature extraction method in point cloud matching. Background technique [0002] In recent years, with the development of science, technology and military aerospace technology, various warfare platforms integrating land, sea, air, space and electricity, light, sound and information have emerged one after another, and information confrontation among various countries has become more and more serious. As it becomes more and more fierce, for future wars, obtaining information becomes the first element of war. Therefore, the data collected by lidar, using 3D point cloud data for point cloud matching, and obtaining the complete 3D information of the target is an indispensable part of information warfare. [0003] With the development of 3D imaging technology, point...

AI Technical Summary

Problems solved by technology

Spin Image descriptors are susceptible to noise interference and are sensitive to non-uniform sampling

PFH description is too time-consuming when computing intensive point clouds. In order to simplify, the FPFH descriptor is proposed, but this descriptor cannot fully display the feature information of all query points.

In the 3DSC and USC descriptors, due to the ambiguity of the azimuth dimension in the calculation of 3DSC, in order to eliminate this shortcoming, the USC descriptor is proposed, but as the noise increases, the descriptor also has certain defects

Images