三维探地雷达道路隐性病害识别方法、系统、介质及产品
By performing in-depth segmentation and multi-dimensional analysis on the three-dimensional ground-penetrating radar data volume, a layered background reference field is constructed, and multi-directional gradients and energy distributions are calculated. This solves the problem of inaccurate identification of hidden road defects in existing technologies and enables accurate identification of defect types and stages.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SICHUAN GEOPHYSICAL SURVEY INST
- Filing Date
- 2026-05-20
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to accurately distinguish and identify the types and development stages of hidden road defects, especially when the defects have complex morphologies in three-dimensional space, leading to inaccurate identification results.
By dividing the vehicle-mounted 3D ground-penetrating radar data volume into multiple layers according to depth, a layered background reference field is constructed. Multi-directional gradient amplitudes are calculated and weighted fusion is performed to generate multi-layer horizontal slices. Geometric morphological parameters are extracted, and the instantaneous amplitude envelope and energy distribution of the vertical profile are calculated. Combining the vertical development mode, tilt characteristics and spatial diffusion characteristics, a multi-dimensional morphological characterization and difference analysis mechanism is established.
It significantly improves the accuracy of identifying hidden road defects, accurately distinguishing between different types and stages of development, and solves the problem of insufficient identification accuracy in existing technologies.
Smart Images
Figure CN122218697B_ABST