Robust hyperspectral image segmentation based on a non-gaussian model

Spectra collected by hyperspectral sensors over samples of the same material are not deterministic quantities. Their inherent spectral variability can be accounted for by making use of suitable statistical models. Within this framework, the Gaussian Mixture Model (GMM) is one of the most widely adopted models for modeling hyperspectral data. Unfortunately, the GMM has been shown not to be sufficiently adequate to represent the statistical behavior of real hyperspectral data, especially for the tails of the distributions. The class of elliptically contoured distributions, which accommodates longer tails, promises to better match the spectral distribution of hyperspectral data. This paper proposes a new Bayesian strategy for learning a non-Gaussian mixture model based on elliptically contoured distributions. This strategy is useful for both fully unsupervised probability density function estimation and clustering purposes. Real hyperspectral imagery is used for experimental evaluation of the proposed strategy. By means of a comparison with GMM Bayesian learning, the proposed method is shown to yield a better fit to hyperspectral data, thus providing a robust technique for clustering hyperspectral images.