Academic Press Library in Signal Processing

Chapter 11

Markov Models and MCMC Algorithms in Image Processing

Xavier Descombes,    Morpheme, Joint Group, INRIA/I3S/IBV, Sophia-Antipolis cedex, France,    [e-mail]

Abstract

Probabilities play a major role in image analysis as they provide general models that take into account the variability of the studied scene and the noise of the sensor. These models are usually expressed through the Bayesian formulation and can thus be divided into a prior and a likelihood. Markov models are particularly interesting as they provide a global formulation through the definition of local function called the potentials. In this chapter, we review the two main family of Markov models, that are the discrete Markov Random Fields and the continuous ...