Multi-Window Detail Enhancement Based Infrared and Visible Image Fusion

The goal of infrared and visible image fusion is to combine complementary information from infrared and visible images of the same scene to generate a high-quality composite image that integrates the advantages of both modalities. Although many current fusion methods achieve satisfactory results, they still suffer from limitations such as insufficient feature resolution extraction from infrared images and inadequate texture information extraction from visible images. This paper proposes a novel fusion method designed to enhance the resolution, detail preservation, and visual consistency of the fused image. The method integrates multi-window detail enhancement with multi-layer residual connections, employing a detail selector and a global feature extractor to separately capture high-frequency and low-frequency features from the infrared and visible images. Experimental results demonstrate that, compared to existing approaches, the proposed method achieves superior fusion quality and better preservation of image details, providing higher-quality data for subsequent image processing tasks.