Abstract
In this paper we present a new algorithm for segmentationof noisy or textured images using the expectation-maximization (EM) algorithm for estimatingparameters of the probability mass function of thepixel class labels and the maximization of the posteriormarginals (MPM) criterion for the segmentation operation.A Markov random field (MRF) model is usedfor the pixel class labels. We present experimental resultsdemonstrating the use of the new algorithm onsynthetic images and medical...

This publication has 8 references indexed in Scilit: