Infrared-Visible Light Image Fusion Method Based on Weighted Salience Detection and Visual Information Preservation

In this paper, we propose a novel image fusion method based on weighted salience detection and visual information preservation. This method is advantageous as the introduced weighted visual salience extraction process effectively highlights the target saliency information in infrared images and integrates this visually significant information into the fused image. To address the issue of overexposure, infrared features are adaptively refined, and a gray-level-driven strategy is employed to fuse these infrared features with visible light images. Our method is tested on multiple image pairs and subjected to qualitative and quantitative evaluations through visual inspection and objective fusion metrics. The results are compared with state-of-the-art fusion techniques. The findings demonstrate that the performance of the proposed method is comparable to or superior to existing methods.