Color shift Model-Based segmentation and fusion for digital autofocusing

摘要: This paper proposes a novel color shift model-based segmentation and fusion algorithm for digital autofocusing of images. The source images are obtained using new multiple filter-aperture configurations. We channels to change the focal point given image at different locations. For each respective location we then select optimal focus information and, finally, use soft decision blending (SDFB) obtain fully-focused proposed consists of: (i) channel shifting alignment varying positions; (ii) region selection sum modified Laplacian (SML); (iii) SDFB, which enables smooth transition across boundaries. By utilizing segmented locations, SDFB can combine with multiple, out-of-focus objects. Experimental results show performance feasibility one or more differently