Pattern selective image fusion for multi-focus image reconstruction

摘要: This paper presents a method for fusing multiple images of static scene and shows how to apply the proposed extend depth field. Pattern selective image fusion provides mechanism combining monochromatic through identifying salient features in source those single fused image. The are first decomposed using filter subtract decimate (FSD) laplacian domain. Thesum-modified-Laplacian (SML) is used obtaining focus images. selected then blended together monotonically decreasing soft decision blending (SDB), which enables smooth transitions across region boundaries. resulting utilizes information that greater than constituent images, while retaining natural verisimilitude. Experimental results show performance extension consumer video camera outputs.