Estimation of mixture and unsupervised segmentation of images

Nicole Marhic, Wojciech Pieczynski

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A statistical approach is presented to the Bayesian unsupervised segmentation of images. The main problem lies in the estimation of a distribution mixture. Iterative conditional estimation provides solutions to such a problem. After a brief review of the general procedure, two stochastic algorithms are described in the case of a finite Gaussian mixture. They are applied to synthetic images corrupted by Gaussian noise in order to estimate the parameters required when performing contextual Bayesian segmentation.

Original languageEnglish
Title of host publicationDigest - International Geoscience and Remote Sensing Symposium (IGARSS)
Editors Anon
PublisherPubl by IEEE
Pages1083-1086
Number of pages4
ISBN (Print)0879426756
Publication statusPublished - 1 Dec 1991
Externally publishedYes
Event1991 International Geoscience and Remote Sensing Symposium - IGARSS'91 - Espoo, Finl
Duration: 3 Jun 19916 Jun 1991

Publication series

NameDigest - International Geoscience and Remote Sensing Symposium (IGARSS)
Volume2

Conference

Conference1991 International Geoscience and Remote Sensing Symposium - IGARSS'91
CityEspoo, Finl
Period3/06/916/06/91

Fingerprint

Dive into the research topics of 'Estimation of mixture and unsupervised segmentation of images'. Together they form a unique fingerprint.

Cite this