Hidden Markov model approach to spectral analysis for hyperspectral imagery

The hidden Markov model (HMM) has been widely used in speech recognition wh ere it models a speech signal as a doubly stochastic process with a hidden state process that can be observed only through a sequence of observations. We present a new application of the HMM in hyperspectral image analysis in spired by the analogy between the temporal variability of a speech signal a nd the spectral variability of a remote sensing image pixel vector. The ide a is to model a hyperspectral spectral vector as a stochastic process where the spectral correlation and band-to-band variability are modeled by a hid den Markov process with parameters determined by the spectrum of the vector that forms a sequence of observations. With this interpretation, a new HMM -based spectral measure, referred to as the HMM information divergence (HMM ID), is derived to characterize spectral properties. To evaluate the perfor mance of this new measure, it is further compared to two commonly used spec tral measures, Euclidean distance (ED) and the spectral angle mapper (SAM), and the recently proposed spectral information divergence (SID). The exper imental results show that the HMMID performs better than the other three me asures In characterizing spectral information at the expense of computation al complexity. (C) 2001 Society of Photo-Optical Instrumentation Engineers.