Multifocus image fusion using local perceived sharpness

Image fusion has become a valuable tool in many applications. In this paper, we propose a conceptually clear and intuitive algorithm for region-based image fusion. Our method considers the properties of multifocus images: in focus regions are clear, but out of focus regions are blurring. Source images are first decomposed into perceptually homogeneous regions and then combine them by use of local perceived sharpness (LPS). Compared to several state-of-the-art image fusion algorithms, experimental results demonstrate that the proposed method is superior in both objective and visual evaluations.