A novel approach for multi-focus image fusion based on SF-PAPCNN and ISML in NSST domain

Abstract

In order to further improve the contrast and sharpness of fused image, a novel multi-focus image fusion algorithm based on spatial frequency-motivated parameter-adaptive pulse coupled neural network (SF-PAPCNN) and improved sum-modified-laplacian (ISML) in nonsubsampled shearlet transform (NSST) domain is proposed in this paper. In its procedural steps, at first, the source images are decomposed into low-frequency and high-frequency components by NSST. The low-frequency components are fused by SF-PAPCNN model, the PAPCNN is designed to estimate the PCNN parameters adaptively according to the input information, and the high-frequency components are fused by ISML model. Finally, the inverse NSST is employed to the fused coefficients to reconstruct the fused image. The superiority of the proposed fusion technique is confirmed by many analytical experimentations on the gray and color multi-focus image data sets. Compared with the state-of-the-art image fusion methods, the proposed fusion algorithm has superior performance in terms of visual inspection and objective evaluation.