A new multi-focus image fusion framework based on focus measures

One of the prime requirements of human visual system and computer vision applications is to acquire a complete image in which focus is on all the objects. Multi-focus image fusion serves as the simplest way to extract and merge complementary information from multiple images into a complete image that has a richer information profile. This work is emphasized to propose a new multi-focus image fusion framework to detect focus regions in the spatial domain. The two focus detection measures based on contrast and edges are utilized by the framework to create the complete image. The proposed framework is evaluated by extensive experiments on various multi-focus images succeeded by a detailed comparative analysis to substantiate the superiority of the proposed work visually and quantitatively.