Hyperspectral Image Enhancement based on sensor simulation and vector decomposition

Hyperspectral Image Enhancement using multispectral data has received considerable attention in recent times in order to achieve higher classification accuracy and more detailed composition analysis. The objective is to obtain an image that has spectral resolution same as that of the hyperspectral image and spatial resolution same as that of multispectral image. While some of the fusion algorithms look at this as a band remapping problem, it is important to maintain the spectral band dependencies in such cases. In this paper, an attempt at using SRFs of different channels is presented to achieve hyperspectral and multispectral image fusion based on vector decomposition. Each multispectral channel fuses detail into only those hyperspectral channels which come into the sensitivity range of that multispectral channel. The results clearly show that the algorithm presented here successfully transfers the spatial details into hyperspectral data while maintaining spectral characteristics of that data.