红外小目标的检测方法、装置、设备及介质
The infrared small target detection method, which combines target neural network and alternating direction multiplier method with matrix broadcasting mechanism, solves the problems of high computational complexity and low detection efficiency in traditional methods, and achieves high-efficiency infrared small target detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG LAB
- Filing Date
- 2026-04-13
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional methods struggle to effectively distinguish between small infrared targets and complex backgrounds, resulting in high computational complexity and low detection efficiency.
A target neural network is used to construct an optimized model for image information extraction and target detection. The alternating direction multiplier method and matrix broadcasting mechanism are used for iterative updating of optimization variables. Parallel computation is performed by combining the matrix broadcasting mechanism of basic image variables and auxiliary optimization variables.
It improves the efficiency of infrared image detection, enhances the contrast between small targets and the background, reduces interference from complex backgrounds, and improves the robustness of detection.
Smart Images
Figure CN122023783B_ABST