Multiband segmentation based on a hierarchical Markov model

Christophe Collet, Fionn Murtagh

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

We develop a new multiscale Markov segmentation model for multiband images. Using quadtree multiple resolution analysis of a multiband image, we use both inter- and intra-scale spatial Markov statistical dependencies. Bayesian inference is used to assess the appropriate number of segments. We exemplify the excellent results which can be obtained with this approach using synthetic images, and in two case studies involving multiband astronomical image sets.

Original languageEnglish
Pages (from-to)2337-2347
Number of pages11
JournalPattern Recognition
Volume37
Issue number12
Early online date20 Jul 2004
DOIs
Publication statusPublished - 1 Dec 2004
Externally publishedYes

Cite this

Collet, Christophe ; Murtagh, Fionn. / Multiband segmentation based on a hierarchical Markov model. In: Pattern Recognition. 2004 ; Vol. 37, No. 12. pp. 2337-2347.
@article{cbce397d345943e78e39863e315a84f6,
title = "Multiband segmentation based on a hierarchical Markov model",
abstract = "We develop a new multiscale Markov segmentation model for multiband images. Using quadtree multiple resolution analysis of a multiband image, we use both inter- and intra-scale spatial Markov statistical dependencies. Bayesian inference is used to assess the appropriate number of segments. We exemplify the excellent results which can be obtained with this approach using synthetic images, and in two case studies involving multiband astronomical image sets.",
keywords = "Bayes factor, Bayes information criterion, Bayesian inference, Generalized Gaussian distribution, Markov random field, Multiband image, Multiresolution, Multiscale, Multispectral image, Quadtree",
author = "Christophe Collet and Fionn Murtagh",
year = "2004",
month = "12",
day = "1",
doi = "10.1016/S0031-3203(04)00190-6",
language = "English",
volume = "37",
pages = "2337--2347",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "12",

}

Multiband segmentation based on a hierarchical Markov model. / Collet, Christophe; Murtagh, Fionn.

In: Pattern Recognition, Vol. 37, No. 12, 01.12.2004, p. 2337-2347.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multiband segmentation based on a hierarchical Markov model

AU - Collet, Christophe

AU - Murtagh, Fionn

PY - 2004/12/1

Y1 - 2004/12/1

N2 - We develop a new multiscale Markov segmentation model for multiband images. Using quadtree multiple resolution analysis of a multiband image, we use both inter- and intra-scale spatial Markov statistical dependencies. Bayesian inference is used to assess the appropriate number of segments. We exemplify the excellent results which can be obtained with this approach using synthetic images, and in two case studies involving multiband astronomical image sets.

AB - We develop a new multiscale Markov segmentation model for multiband images. Using quadtree multiple resolution analysis of a multiband image, we use both inter- and intra-scale spatial Markov statistical dependencies. Bayesian inference is used to assess the appropriate number of segments. We exemplify the excellent results which can be obtained with this approach using synthetic images, and in two case studies involving multiband astronomical image sets.

KW - Bayes factor

KW - Bayes information criterion

KW - Bayesian inference

KW - Generalized Gaussian distribution

KW - Markov random field

KW - Multiband image

KW - Multiresolution

KW - Multiscale

KW - Multispectral image

KW - Quadtree

UR - http://www.scopus.com/inward/record.url?scp=12344270152&partnerID=8YFLogxK

U2 - 10.1016/S0031-3203(04)00190-6

DO - 10.1016/S0031-3203(04)00190-6

M3 - Article

VL - 37

SP - 2337

EP - 2347

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 12

ER -