Image fusion method and robustness test based on multiscale decomposition and spiking cortical model

Abstract

The main purpose of this paper is to find an efficient image fusion algorithm for multifocus images, based on the shift-invariance, multi-scale and multi-directional properties of nonsubsampled contourlet transform (NSCT) along with human visual characteristics of spiking cortical model (SCM). Firstly, maximum selection rule (MSR) is used to fuse low frequency coefficients of NSCT. Secondly, the time matrix of SCM is set as criteria to select coefficients of high frequency subband. The effectiveness of the proposed algorithm is achieved by the comparison with existing fusion methods. Objective tests and analysis conducted under different noised source image environments proved the robustness of the proposed fusion method.