一种基于交叉聚焦线性注意力的红外和可见光图像融合方法
The image fusion method using cross-focusing linear attention solves the problems of long-distance dependence and insufficient cross-modal feature interaction in infrared and visible light image fusion, achieving efficient information preservation and detail enhancement, and is suitable for applications such as night vision, search and rescue, perimeter protection and intelligent transportation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- KUNMING UNIV OF SCI & TECH
- Filing Date
- 2025-08-02
- Publication Date
- 2026-07-17
AI Technical Summary
Existing infrared and visible light image fusion methods fail to effectively model long-range dependencies and cross-modal feature interactions, resulting in the loss of detailed information in the fused image.
An image fusion method based on cross-focusing linear attention is adopted, which includes a four-level encoder and decoder. Through bi-branch feature extraction, adaptive feature correction, cross-focusing linear attention fusion and frequency-aware feature aggregation, combined with a joint loss function optimization model, it achieves full interaction and information preservation of cross-modal features.
It significantly improves the visual quality and robustness of fused images, better preserves thermal targets and visible light details, enhances the semantic richness and contrast of images, and is suitable for applications such as night vision, search and rescue, perimeter protection, and intelligent transportation.
Smart Images
Figure CN120976036B_ABST