Bayesian morphology: fast unsupervised Bayesian image analysis.

Journal of the American Statistical Association, June, 1999 by Forbes, Florence; Raftery, Adrian E.

A new procedure for using Bayesian statistics in the segmentation, classification and restoration of images has been developed. The technique, called Bayesian morphology, is based on mathematical morphology and the ICM method. The idea behind this approach is that the ICM algorithm is equivalent to a form of mathematical morphology when the original image is discrete and the transmission noise can be simulated by a Potts model. It is also possible to carry out maximum likelihood estimation of the parameters using the novel scheme.

1. INTRODUCTION

We consider the problem of image segmentation or classification, or of image restoration when the true scene is made up of a small number of unordered colors. Here we consider only the generic version of this problem,...

Read the rest of this article with a Free Trial at HighBeam Research.

Premium Content Partnership | HighBeam Research provides an in-depth online archive library of reference works.