Precise comparison and analysis method for special-shaped curved surface component based on 3D laser scanning
By performing primary and secondary clustering on the wheel hub point cloud, the objective function of the ICP algorithm is optimized, solving the problem that the ICP algorithm ignores the geometric features of the wheel hub, and achieving high accuracy and high reliability of wheel hub point cloud registration.
CN122048930BActive Publication Date: 2026-07-21CHONGQING HONGCHUAN MACHINERY MANUFACTURING CO LTD
View PDF 2 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING HONGCHUAN MACHINERY MANUFACTURING CO LTD
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-21
AI Technical Summary
Technical Problem
In existing technologies, the ICP algorithm ignores the geometric features of the wheel hub when performing point cloud registration, resulting in poor detection reliability.
Method used
A 3D laser scanning-based method is used to perform primary and secondary clustering on the wheel hub point cloud. By obtaining the density characteristics and local curvature of the wheel hub category point cloud clusters, the objective function of the ICP algorithm is optimized, taking into account the spatial location and local geometric feature differences of the point cloud.
Benefits of technology
It significantly improves the accuracy and reliability of hub point cloud registration, and enhances the precision and reliability of detection.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122048930B_ABST
Abstract
The present application relates to the technical field of data processing, in particular to a kind of special-shaped curved surface component accurate contrast analysis method based on 3D laser scanning.First, according to the density feature, point cloud is clustered once, obtains multiple first point cloud clusters, realizes unsupervised classification;Second, by calculating the error of actual and predicted overall curvature under different number of neighbor points, the fluctuation of the overall curvature of each first point cloud cluster can be quantified, the first point cloud cluster is clustered twice, and the hub category point cloud cluster with different geometric features is identified according to the fluctuation of the overall curvature, which improves the accuracy of identification;Then, by obtaining the optimal number of neighbor points of each type of hub category point cloud cluster and used for surface fitting, the stability of geometric feature measurement is improved, so that the final distance measurement integrates spatial position and local geometric information;Finally, by optimizing the objective function and threshold judgment, the point cloud registration accuracy and the reliability of hub contrast analysis are significantly improved.
Need to check novelty before this filing date? Find Prior Art