Multiresolution image segmentation

Author

Comer, Mary L. ; Delp, Edward J.

Author_Institution

Sch. of Electr. Eng., Purdue Univ., West Lafayette, IN, USA

Volume

4

fYear

1995

fDate

9-12 May 1995

Firstpage

2415

Abstract

In this paper we present a new algorithm for segmentation of noisy or textured images using a multiresolution Bayesian approach. Our algorithm is different from previously proposed multiresolution segmentation techniques in that we use a multiresolution Gaussian autoregressive (AR) model for the pyramid representation of the observed image. Our algorithm also approximates the “maximization of the posterior marginals” (MPM) estimate of the pixel class labels at each resolution, from coarsest to finest, unlike previously proposed techniques, which have been based on MAP estimation. Experimental results are presented to demonstrate the performance of the new algorithm

Keywords

Bayes methods; Gaussian processes; autoregressive processes; image resolution; image segmentation; image texture; parameter estimation; image segmentation; maximization of the posterior marginals estimate; multiresolution Bayesian approach; multiresolution Gaussian autoregressive model; noisy images; pixel class labels; pyramid representation; textured images; Bayesian methods; Cost function; Image resolution; Image segmentation; Lattices; Parameter estimation; Pixel; Random variables; Spatial resolution; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1995. ICASSP-95., 1995 International Conference on

Conference_Location

Detroit, MI

ISSN

1520-6149

Print_ISBN

0-7803-2431-5

Type

conf

DOI

10.1109/ICASSP.1995.479980

Filename

479980