Non-Gaussian Linear Mixing Models for Hyperspectral Images

Modeling of hyperspectral data with non-Gaussian distributions is gaining popularity in recent years. Such modeling mostly concentrates on attempts to describe a distribution, or its tails, of all image spectra. In this paper, we recognize that the presence of major materials in the image scene is likely to exhibit nonrandomness and only the remaining variability due to noise, or other factors, would exhibit random behavior. Hence, we assume a linear mixing model with a structured background, and we investigate various distributional models for the error term in that model. We propose one model based on the multivariate t-distribution and another one based on independent components following an exponential power distribution. The former model does not perform well in the context of the two images investigated in this paper, one AVIRIS and one HyMap image. On the other hand, the latter model works reasonably well with the AVIRIS image and very well with the HyMap image. This paper provides the tools that researchers can use for verifying a given model to be used with a given image