Research on multi-focus image fusion algorithm based on quadtree

Abstract

Most of the multi-focus fusion algorithms currently are prone to image blur and loss of detail information. This paper proposes a multi-focus fusion algorithm based on quadtree decomposition which can almost overcome the above shortcomings. The research on multi-focus fusion algorithm based on quadtree decomposition is to divide the original image into several image sub-blocks and check regional consistency for each block to obtain the optimal block of the source image, and then detect the focus area for each block to obtain the initial fused decision image. Finally, the fused decision image is subjected to perform morphological processing to obtain a final fused image. Through extensive experiments on different source images, we show that the proposed method has better adaptability.