Entropy powered image fusion based on multi scale top-hat transform

An entropy powered image fusion algorithm using multi scale top-hat transform is proposed in this paper. At the beginning, multi scale white and black top-hat transforms are used to extract bright and black image regions of images. To be efficient for image fusion, an entropy powered image fusion strategy is proposed. The extracted bright and black image regions are firstly powered corresponding to each pixel, then the powered bright and black image regions are fused in the fusion image, which well preserves the useful image information in the final fusion image. Experimental results using multi modal images and multi focus images show that, our algorithm is efficient and could be widely used in the field of image fusion.