Low-overlapping-rate three-dimensional point cloud registration method

Abstract

The invention discloses a low-overlapping-rate three-dimensional point cloud registration method, and belongs to the technical field of machine vision. The method aims at solving the problems of highregistration difficulty, low precision and the like of two point clouds with low overlapping rate. The method comprises the following steps: firstly, establishing a multi-scale descriptor by utilizingcurvature characteristics of point cloud to ensure that point cloud data is complete and redundant data is minimum; secondly, performing corresponding relation clustering partitioning by utilizing the angle difference of the multi-scale descriptors to obtain an overlapping region of the source point cloud and the target point cloud; and finally, substituting the point clouds of the overlapping region and the corresponding relationship of the point clouds into a convex optimization problem, removing outliers and optimizing the corresponding relationship, realizing coarse registration, and refining by utilizing an ICP algorithm. According to the invention, the useful search range of point cloud registration can be reduced, the registration calculation amount is reduced, and more advantageous registration precision and time efficiency are provided for point cloud data with a low initial overlapping degree. The method can be widely applied to the fields of three-dimensional model reconstruction, cultural heritage management, robot navigation and positioning and the like.

Description

technical field

[0001] The invention belongs to the technical field of machine vision, and in particular relates to a three-dimensional point cloud registration method with a low overlap rate. Background technique

[0002] In recent years, 3D point cloud data has been widely used in 3D model reconstruction, cultural heritage management, robot navigation and positioning and other fields. Fast and accurate 3D point cloud registration technology is the key technology and research focus. The goal of 3D point cloud registration is to find the optimal rigid body transformation that aligns two input point clouds to a common coordinate system. In practical applications, the data may be heavily occluded, and the overlapping area between two point clouds is small, which makes the process of finding the optimal rigid body transformation challenging. Therefore, finding fast, accurate, and robust registration algorithms for point clouds with smaller overlapping ranges is an active rese...

Images