VMDM-fusion: a saliency feature representation method for infrared and visible image fusion

A VGG-based fusion method (named VMDM-Fusion) that employs multiple decision maps is proposed to fuse infrared and visible images. Our method first feeds the infrared and visible images into a pre-trained model of VGG-16 to extract the features. Then, a feature representation method we designed uses these features to construct saliency maps. Next, these maps, in combination with a guided filter, are used to construct multiple decision maps. Lastly, the final fused image is obtained by weighting the source images based on the multiple decision maps. This is the first time a decision map is introduced in the field of infrared and visible image fusion. The experimental results demonstrate that the proposed method outperforms state-of-the-art infrared and visible image fusion methods.