Publications

Abstract : Pansharpening refers to fusion of a multispectral (MS) image with a panchromatic (PAN) image to obtain an image with both high spatial and spectral resolution. Among the many existing pansharpening methods the most widely used is the component substitution method. In this paper an improved model of this existing method is proposed which is based on the matting model and an adaptive fusion technique known as the harmony search algorithm. In this model the MS image which can be decomposed to three components via, the alpha channel, spectral foreground and background is used to reconstruct a high resolution MS image by substituting the PAN image instead of the alpha channel. The fusion is done using an adaptive technique, the harmony search which finds the optimal weight value for effective fusion and it is observed that this method gives a better value for many quantitative indices.