一种用于隧道内突水突泥灾害检测的方法及系统
By using UAV cross-section recognition and YOLOv8 network processing for thermal imaging and field-of-view images, diffuse reflection interference within the tunnel is identified and suppressed, enabling efficient and accurate detection of water and mud inrush disasters within the tunnel.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV
- Filing Date
- 2025-09-05
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for detecting water and mud inrush disasters in tunnels suffer from low image acquisition quality, making it difficult to distinguish between real disasters and reflected noise, especially under uneven lighting conditions where the detection effect is unsatisfactory.
UAVs are used for tunnel cross-section identification, combined with thermal imaging and field-of-view image acquisition. The YOLOv8-derived convolutional network RefNet is used to identify and suppress diffuse reflection interference. Feature extraction and fusion are performed through a cross-modal adaptive fusion attention mechanism. The YOLOv8 neck network is used for disaster assessment.
It improves the accuracy of detecting water and mud inrush disasters in tunnels, reduces the false detection rate, adapts to complex tunnel environments, and maintains high-quality image acquisition.
Smart Images
Figure CN120976808B_ABST