Research On Energy-efficient Multi-focus Image Fusion Algorithm Based On Morphology

With the development of sensor technology that brings the abundance of image sources,image fusion technique has been applied widely in many different fields,such as military affairs,remote sensing,computer vision,medical imaging and so on,and also has been a hot research topic at home and abroad.Recently,image fusion technique always has been developing rapidly with the multi-scale analysis tools improved continually,which certainly leads to better fusion performance.However,at the meanwhile,the computational complex of fusion algorithms also rise sharply,which,as a result,limits them to be applied at the application environment with limited energy and high requirement for real-time performance,such as wireless multi-media sensor networks.Now,there are still many problems needed to be solved for fusion algorithms based on traditional multi-scale analysis tools,such as the discontinuity of focus regions,distorted brightness,poor robustness.Therefore,how to decrease the complex of fusion algorithm with keeping great fusion performance is still a huge challenge.This paper mainly focuses on the research of low computational complex multi-focus image fusion algorithm.The main contributions of this dissertation are summarized as follows:(1)According to the advantages of the strong identification ability and low computational complex of multi-scale neighbor distance analysis tool,an energy-efficient multi-focus image fusion algorithm based on neighbor distance and mathematical morphology technique is proposed.The proposed method effectively overcomes the problems of both the discontinuity of focus region and distorted brightness.The method to detect focus areas with block scheme in frequent domain ensures lower computational complex for the proposed method.(2)According to the theory that grey relational analysis has unique superiority in processing small simple data,an energy-efficient multi-focus image fusion algorithm based on grey relation of similarity and mathematical morphology technique is proposed.This method has low computational complex while keeping the continuity of focus region very well.The proposed algorithm has great robustness,because it can always realize good fusion performance in fusing clean multi-focus images.mis-registration multifocus images or noisy multi-focus images.