The goal of fusing multi-focus images is to obtain one image that has all the significant features from each input. The fusion process is needed because of the limitations of the optical lens depth of field that is used to capture images, so images of various focused regions are produced. In this paper, a multi-focus image fusion algorithm is proposed. It is based on utilizing the biorthogonal wavelet transform to extract details and edges from the input images by making the approximation subband equal zero and applying an inverse transform to get images that have only edges, lines, and details. The average gradient metric, which is used to represent sharpness and clarity, is calculated as an activity measurement for each NxN block of the resulted edge images and is used to merge the corresponding blocks of the multi-focus input images to produce the fused one. Consistency verification is used to improve the fusion process. The performance of the proposed method was evaluated and contrasted with a number of other state-of-the-art fusion methods. Experimental results clearly show that the suggested approach is a feasible and efficient multi-focus imaging technique. .
