一种基于三维点云的结构面智能识别方法
By calculating point cloud normal vectors and using the DBSCAN algorithm for clustering, the problems of complexity in structural surface recognition and noise point processing in existing technologies are solved, achieving simple and accurate structural surface recognition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- INST OF ROCK & SOIL MECHANICS CHINESE ACAD OF SCI
- Filing Date
- 2024-01-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing structural surface recognition methods based on 3D point clouds suffer from problems such as complex preprocessing, strong subjectivity, inability to remove noise points, and complex usage.
The point cloud normal vectors are calculated using K-nearest neighbor search and least squares method. The DBSCAN algorithm is used for clustering to automatically adjust the consistency of the normal vector direction and remove noise points. Joint groups and structural surfaces are obtained through DBSCAN clustering, and the attitude is calculated using area-weighted average.
The preprocessing process is simplified, subjectivity is reduced, noise points are automatically removed, the recognition process is simple, and the results are objective and accurate.
Smart Images
Figure CN118072300B_ABST