An adaptive remote sensing image fusion algorithm based on directionlet transform

Abstract

A new adaptive remote sensing image fusion algorithm based on Directionlet transform is proposed in this paper. Directionlet transform introduces multi-directional anisotropic basic function based on integer lattice which can effectively capture the anisotropic characteristics of images. For the high-frequency multi-directional coefficients, a contrast-based criterion is adopted and the low-pass coefficients are then selected adaptively based on a weighted criterion. Experimental results show that the proposed remote sensing fusion algorithm outperforms the wavelet-based fusion algorithm, both in terms of visual effect as well as quantized objective evaluation index, which proves that this algorithm is a feasible and effective remote sensing image fusion method.