Infrared and visible image fusion using multi-scale edge-preserving decomposition and multiple saliency features

The fusion technique of infrared and visible images is broadly adopted in many computer vision tasks, such as pattern recognition, target detecting, tracking, and surveillance. However, many commonly used fusion methods usually ignore the visual naturalness and information fidelity of the fused image, which make the fused image unsuitable for human visual perception. To address these defects of existing methods, this work presents a new fusion method for a well-pleasing visual effect. Firstly, multi-scale decomposition using bi-exponential edge-preserving smoother is presented to decompose the source image into multi-scale representations, aiming to extract the multi-scale structure information and refrain from the halos around the edges. Secondly, a 'weighted average' rule is used to merge the base layers, which can enhance the contrast and highlight the targets. And a 'multiple saliency features' rule is proposed to merge the detail layers with the goal of acquiring the detail and texture information, which can provide rich details for the fused images with high visual information fidelity. Finally, reconstruction for the final image is carried out. Extensive experiment results effectively demonstrate that the proposed method achieves better performance than several state-of-the-art fusion methods in naturally visual effect and retaining edge details.