Multi-focus image fusion using energy of image gradient and gradual boundary smoothing

Existing imaging cameras usually have limited depth of field, so it is not possible to obtain an image in which all the objects are focused. To obtain an all-in-focus image, this paper proposes a multi-focus image fusion algorithm using energy of image gradient and gradual boundary smoothing. The initial fused image is generated using energy of image gradient which determines the sharpest pixels of the source images. Since the energy of image gradient in smooth regions is inaccurate, the sharpest pixels selection is performed on edge regions. The selection of the sharpest pixels in smooth regions is decided by the selection of the nearest edge region. The initial fused image contains visual artifact in boundary region which is fused with one side of the region from first source image and the other side of the region from another source image. To suppress this undesired artifact, the proposed method uses gradual boundary smoothing. Experimental results show that the proposed algorithm is efficient.