Hyperspectral image processing and analysis

Hyperspectral data pose challenges to image interpretation, because of the need for calibration, redundancy in information, and high data volume due to large dimensionality of the feature space. In this article, a general framework is presented for working with hyperspectral imagery, including removal of atmospheric effects, imaging spectroscopy, dimensionality reduction and classification of imagery. The phenomenon of mixture modelling is briefly discussed, followed by a recent development in mapping the classes at sub-pixel level based on the principle of super-resolution