Bayes factors for edge detection from wavelet product spaces

Fionn Murtagh, Jean Luc Starck

Research output: Contribution to journalReview article

7 Citations (Scopus)

Abstract

Interband wavelet correlation provides one approach to defining edges in an image. Interband wavelet products follow long-tailed density distributions, and in such a context thresholding is very difficult. We show how segmentation using a Markov-field spatial dependence model is a more appropriate approach to demarcating edge and non-edge regions. A key part of this work is quantitative assessment of goodness of edge versus nonedge fit. We introduce a formal assessment framework based on Bayes factors. A detailed example is used to illustrate these results.

LanguageEnglish
Pages1375-1382
Number of pages8
JournalOptical Engineering
Volume42
Issue number5
DOIs
Publication statusPublished - 1 May 2003
Externally publishedYes

Fingerprint

edge detection
Edge detection
products
density distribution

Cite this

Murtagh, Fionn ; Starck, Jean Luc. / Bayes factors for edge detection from wavelet product spaces. In: Optical Engineering. 2003 ; Vol. 42, No. 5. pp. 1375-1382.
@article{0fba7ce6b03f4fae8065e107a64fd684,
title = "Bayes factors for edge detection from wavelet product spaces",
abstract = "Interband wavelet correlation provides one approach to defining edges in an image. Interband wavelet products follow long-tailed density distributions, and in such a context thresholding is very difficult. We show how segmentation using a Markov-field spatial dependence model is a more appropriate approach to demarcating edge and non-edge regions. A key part of this work is quantitative assessment of goodness of edge versus nonedge fit. We introduce a formal assessment framework based on Bayes factors. A detailed example is used to illustrate these results.",
keywords = "Bayes factor, Bayes information criterion, Edge detection, Heavy-tailed distribution, Likelihood, PLIC, Pseudolikelihood information criterion, Segmentation, Wavelet transform",
author = "Fionn Murtagh and Starck, {Jean Luc}",
year = "2003",
month = "5",
day = "1",
doi = "10.1117/1.1564104",
language = "English",
volume = "42",
pages = "1375--1382",
journal = "Optical Engineering",
issn = "0091-3286",
publisher = "SPIE",
number = "5",

}

Bayes factors for edge detection from wavelet product spaces. / Murtagh, Fionn; Starck, Jean Luc.

In: Optical Engineering, Vol. 42, No. 5, 01.05.2003, p. 1375-1382.

Research output: Contribution to journalReview article

TY - JOUR

T1 - Bayes factors for edge detection from wavelet product spaces

AU - Murtagh, Fionn

AU - Starck, Jean Luc

PY - 2003/5/1

Y1 - 2003/5/1

N2 - Interband wavelet correlation provides one approach to defining edges in an image. Interband wavelet products follow long-tailed density distributions, and in such a context thresholding is very difficult. We show how segmentation using a Markov-field spatial dependence model is a more appropriate approach to demarcating edge and non-edge regions. A key part of this work is quantitative assessment of goodness of edge versus nonedge fit. We introduce a formal assessment framework based on Bayes factors. A detailed example is used to illustrate these results.

AB - Interband wavelet correlation provides one approach to defining edges in an image. Interband wavelet products follow long-tailed density distributions, and in such a context thresholding is very difficult. We show how segmentation using a Markov-field spatial dependence model is a more appropriate approach to demarcating edge and non-edge regions. A key part of this work is quantitative assessment of goodness of edge versus nonedge fit. We introduce a formal assessment framework based on Bayes factors. A detailed example is used to illustrate these results.

KW - Bayes factor

KW - Bayes information criterion

KW - Edge detection

KW - Heavy-tailed distribution

KW - Likelihood

KW - PLIC

KW - Pseudolikelihood information criterion

KW - Segmentation

KW - Wavelet transform

UR - http://www.scopus.com/inward/record.url?scp=0038718950&partnerID=8YFLogxK

U2 - 10.1117/1.1564104

DO - 10.1117/1.1564104

M3 - Review article

VL - 42

SP - 1375

EP - 1382

JO - Optical Engineering

T2 - Optical Engineering

JF - Optical Engineering

SN - 0091-3286

IS - 5

ER -