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.