A method and system for restoring degraded infrared images
By constructing different types of degraded images and training a super-resolution processing model, the problem of poor infrared image restoration under severe weather conditions was solved, achieving high-quality restoration and enhanced model stability, making it suitable for edge deployment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN GUIDE SENSMART TECH CO LTD
- Filing Date
- 2025-12-18
- Publication Date
- 2026-05-26
AI Technical Summary
Existing technologies are not effective in restoring infrared images under adverse weather conditions, especially in rain, snow, fog, and strong winds, where image quality deteriorates. Furthermore, existing models require a large amount of computation and are not suitable for edge deployment.
By constructing different types of degraded images and combining them with super-resolution processing models for targeted training, various image degradation factors, including atmospheric scattering, turbulence, and infrared sensor noise, are simulated, thereby improving the robustness and stability of the model.
Achieving high-quality infrared image restoration under diverse weather conditions enhances the model's robustness and computational efficiency, making it suitable for edge deployment.
Smart Images

Figure CN122089608A_ABST