Detail enhanced multi-source fusion using visual weight map extraction based on multi scale edge preserving decomposition

Image fusion is an important problem in image analysis. The abundant detail information in multisource images should be preserved or even enhanced for the fused result. In this paper, using visual weight map extraction based on multiscale edge preserving decomposition, a details enhancing fusion algorithm is proposed. The saliency extraction based visual weight map is used for image fusion at different scales, which is decomposed by edge preserving image smooth approach. With this method, the details information which human visual system interests can be preserved and enhanced easily. The quantitative comparisons of experiments demonstrate that the proposed algorithm performs better than other methods, especially in visual effect and keeping details.