Markov random measure fields for image analysis

Author

Marroquín, José L. ; Arce, Edgar ; Botello, Salvador

Author_Institution

Centro de Investigaciones en Matematicas, Guanajuato, Mexico

Volume

1

fYear

2002

fDate

2002

Abstract

A new Bayesian formulation for the image segmentation problem is presented. It is based on the key idea of using a doubly stochastic prior model for the label field, which allows one to find exact optimal estimators by the minimization of a differentiable function. Comparisons with existing methods on synthetic images are presented, as well as realistic applications to the segmentation of magnetic resonance volumes, to motion segmentation, and to edge-preserving filtering.

Keywords

Bayes methods; Markov processes; image segmentation; minimisation; parameter estimation; Bayesian formulation; Markov random fields; Markov random measure fields; differentiable function minimization; doubly stochastic prior model; edge-preserving filtering; exact optimal estimators; image analysis; image segmentation; label field; magnetic resonance volumes; motion segmentation; synthetic images; Bayesian methods; Computer vision; Image analysis; Image edge detection; Image motion analysis; Image segmentation; Magnetic field measurement; Magnetic resonance; Motion segmentation; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing. 2002. Proceedings. 2002 International Conference on

ISSN

1522-4880

Print_ISBN

0-7803-7622-6

Type

conf

DOI

10.1109/ICIP.2002.1038137

Filename

1038137