Low-overlap weak-feature three-dimensional measurement point cloud fine registration method with introduction of plane constraint

By combining a binocular structured light camera and a robotic arm, along with KD-tree and RANSCA algorithms, a planar constraint point cloud fine registration method was constructed. This method solved the problem of accurate registration of point clouds with low overlap rates for large, weak feature components, and achieved efficient 3D measurement results.

CN116245921BActive Publication Date: 2025-11-11HUAZHONG UNIV OF SCI & TECH +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202211642934.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-12-20
Publication Date
2025-11-11
Estimated Expiration
2042-12-20

Smart Images

  • Figure CN116245921B_ABST
    Figure CN116245921B_ABST
Patent Text Reader

Abstract

The application belongs to the technical field of three-dimensional measurement, and particularly discloses a low-overlap weak feature three-dimensional measurement point cloud fine registration method with plane constraint, which comprises the following steps: obtaining surface three-dimensional point cloud data of a large weak feature component through measurement, and performing coarse registration on the measurement point cloud data by using an end transformation matrix of a mechanical arm; finding the nearest point of the coarse registration source point cloud in the target point cloud, constructing a point cloud pair error probability distribution, screening effective point pairs based on the error probability distribution, and finding an effective overlap area; finding the plane structure of the effective overlap area of the source point cloud and the target point cloud respectively by using the RANSCA algorithm, and dividing the effective point cloud pair into a plane area and a non-plane area; projecting the source point cloud of the plane area to the target point cloud plane area, constructing a point cloud fine registration cost function with plane constraint on the basis of the ICP algorithm, and iteratively optimizing to obtain a point cloud transformation matrix with accurate registration. The application is accurate in registration and suitable for the registration of low-overlap point clouds.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Automatic registration method for three-dimensional point cloud data

    CN106780459A

  • An automatic registration method and apparatus for point cloud data based on planar features

    CN110443836B

  • A Point Cloud Stitching Method and System Based on Weak Feature Neighborhood Probability Matching

    CN110992259B

  • Point cloud high-precision automatic registration method based on biological vision

    CN111429494A

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

    CN112150523A