通过线路故障检测模型检测输电线路故障的方法及设备
By extracting and fusing features from visible light and infrared images using YOLOv5 and a dual-stream backbone network, the efficiency and accuracy issues of transmission line fault detection in existing technologies are resolved, enabling efficient and accurate fault identification in various environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- YANTAI STATE GRID ZHONGDIAN ELECTRIC CO LTD
- Filing Date
- 2025-07-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for detecting faults in transmission lines suffer from problems such as low inspection efficiency, high false positive rate, and insufficient accuracy. They are particularly difficult to accurately identify early faults in harsh environments, and different spectral detection methods have their own limitations and cannot fully cover the line status.
YOLOv5 is used as the basic architecture, and a dual-stream backbone network is used to extract and fuse features from visible light and infrared images. The first backbone network extracts multi-scale feature maps of visible light images, and the second backbone network extracts multi-scale feature maps of infrared images. Multiple feature fusions are performed in the neck network. Finally, the detection results are output through the prediction module, and the model is optimized using the SIoU loss function.
It improves the accuracy and precision of transmission line fault detection, effectively identifies line faults under different lighting and environmental conditions, reduces misjudgments, and enhances the comprehensiveness and reliability of detection.
Smart Images
Figure CN120876400B_ABST