A novel weight fusion approach for multi-focus image based on NSST transform domain

Abstract

In this paper, a novel weight multi-focus image fusion method based on the non-subsampled shearlet transform (NSST) is proposed. Firstly, the NSST is employed to decompose the source image into low frequency and high frequency sub-band coefficients with different-scale and different-directional; Secondly, the local least root mean square error (RMSE) is used as a new fusion weight for the fusion of low frequency sub-band coefficients; In order to effectively preserve the edges of the image, a novel fusion rule based on edge preserving weight is proposed to fuse high frequency sub-band coefficients. Finally, the fused image can be obtained through the inverse NSST on the fused coefficients. The experimental results show that the fused image not only contains rich details, but also preserves the notable structural features. Compare with the current state-of-the-art fusion methods, it produces better visual effect and quality assessment.