Infrared and visible image fusion based on fourth-order partial differential equations and sub-window variance filters

Image fusion aims to combine images of different modalities in a specific way so that the images have the properties of different modalities. Fusion of infrared and visible images enables the fused image to contain not only rich details but also thermal information. We propose an image fusion method based on fourth-order partial differential equations (FPDE) and sub-window variance filters (SVF). First SVF is utilized on the source images to remove noise. Subsequently, the images are decomposed two distinctive methodologies to obtain detail and base layers. To acquire a fine-grained texture, the primary layer of image decomposition is executed using FPDE. Then SVF with different parameters is performed on the images to get coarse-grained edges. The detail and base layers are subsequently fused with different strategies. Experimental results demonstrate that our proposed method surpasses the performance of current state-of-the-art fusion methods.