High quality multi-focus image fusion using self-similarity and depth information

Due to the limited depth of field in a camera, some imaging objects will be blurred if they are located far from the focus plane and the other objects on the plane will be clear. Multi-focus image fusion synthesizes a sharp image from multiple partially focused images. However, traditional fused images usually suffer from blurring effects and pixel distortions. In this paper, we explore two unique characteristics of multi-focus images: (1) The self-similarity of a single image and the shared similarity among multiple source images; (2) The distances from object to focal plane. The former characteristic is used to identify image structure-driven regions while the latter refine the image clarity by automatically estimating depth information of blurred images. Experimental results demonstrate that the proposed method outperforms the state-of-the-art fusion methods on image quality and objective fusion criteria.