一种基于图像和点云融合的隧道节理识别方法
The tunnel joint identification method, which integrates image and point cloud data, solves the problems of low efficiency and poor accuracy in traditional tunnel joint identification. It achieves high-precision joint identification and parameter extraction in complex environments and supports health monitoring throughout the entire life cycle of tunnels.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TONGJI UNIV
- Filing Date
- 2026-04-24
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional tunnel joint identification methods are labor-intensive, inefficient, and subject to environmental limitations, resulting in poor data accuracy and consistency. Existing single-modal identification methods are costly and time-consuming, making it difficult to meet the high-precision requirements during the construction period.
A tunnel joint recognition method based on image and point cloud fusion is adopted. Multi-source data is acquired synchronously, preprocessed, and then a deep learning network is used for joint detection and segmentation. Combined with adaptive density clustering and spatial matching of point cloud, a high-confidence three-dimensional joint point set is obtained, and plane fitting is performed to extract geometric parameters.
It enables continuous three-dimensional reconstruction and parametric measurement of joints in complex environments, improving identification accuracy and engineering usability, providing comprehensive quantitative data support, and is applicable to tunnel construction and operation and maintenance.
Smart Images
Figure CN122089737B_ABST