Next: Finite Mixture Model
Up: Segmentation of Brain MR
Previous: Introduction
Hidden Markov Random Field Model
Let
and
be two alphabets:
Let
be the set of indices and
denote any family of random variables
indexed by
,
in which each random variable Ri takes
a value zi in its state space. Such a family R is called a
random field. The joint event
is
simplified to R=r where
is a
configuration of R, corresponding to a realization of
this random field. Let X and Y be two such random fields whose
state spaces are
and
respectively so
that for
we have
and
.
Let x denote a configuration
of X and
be the set of all possible configurations
so that
Similarly, let y be a configuration of Y and
be the set of all possible configurations so that
Given
,
Yi follows a conditional probability
distribution
 |
(1) |
where
is the set of parameters. For all
,
the
function family
has the same known analytic
form. We also assume that (X, Y) is pairwise independent,
meaning
 |
(2) |
In order to develop the HMRF model we firstly take the standard FM
model as a comparison.
Next: Finite Mixture Model
Up: Segmentation of Brain MR
Previous: Introduction
Yongyue Zhang
2000-05-11