Implementation of Nonlocal Pansharpening Image Fusion

Abstract

This paper focuses on the implementation of the pansharpened image fusion technique proposedin the companion paper [A Nonlocal Variational Model for Pansharpening Image Fusion, SIAMJournal on Imaging Sciences, 2014, to appear]. Pansharpening refers to the process of inferringa high resolution multispectral image from a high resolution panchromatic image and low resolution multispectral one. Although quite successful in terms of relative error, state-of-the-artpansharpening methods still introduce relevant color artifacts. The variational pansharpeningmodel proposed by Duran et al. incorporates a nonlocal regularization term that takes ad-vantage of image self-similarity, leading to signicant reduction of the above-mentioned colorartifacts.