Infrared And Visible Image Fusion Based On Spatial Convolution Sparse Representation

In the traditional sparse representation based infrared and visible image fusion method, the detail information is not able to be effectively extracted, resulting in the decrease of infrared target intensity and the blurry of visible background information. In order to solve the above issue, a new image fusion method based on spatial convolution sparse representation is proposed. Firstly, a spatial convolution sparse representation is used to perform two-scale decomposition of infrared and visible images by introducing a gradient regularization, and the detail and intensity information are extracted effectively from the source images. Then, the weighted average rule and the maximum selection rule are used to fuse the base layer and detail layer images, respectively. Finally, the fusion image is constructed. Experimental results illustrate that the proposed method is superior to the traditional fusion method based on sparse representation.