Bayesian Image Classification with Baddeley’s Delta Loss

Arnoldo Frigessi; Håvard Rue

doi:10.1080/10618600.1997.10474727

Bayesian Image Classification with Baddeley’s Delta Loss

Arnoldo Frigessi, Håvard Rue^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

9 Scopus citations

Abstract

In this article we adopt Baddeley's delta metric as a loss function in Bayesian image restoration and classification. We develop a new algorithm that allows us to approximate the corresponding optimal Bayesian estimator. With this algorithm good practical estimates can be obtained at approximately the same computational cost as traditional estimators like the marginal posterior mode (MPM). A comparison of our proposed classification with MPM shows significant advantages, especially with respect to fine structures.

Original language	English (US)
Pages (from-to)	55-73
Number of pages	19
Journal	JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
Volume	6
Issue number	1
DOIs	https://doi.org/10.1080/10618600.1997.10474727
State	Published - Mar 1997
Externally published	Yes

Keywords

Asymmetric loss functions
Bayesian inference
Distance between binary images
Image restoration
Markov chain Monte Carlo methods
Metropolis algorithm

ASJC Scopus subject areas

Discrete Mathematics and Combinatorics
Statistics and Probability
Statistics, Probability and Uncertainty

Access to Document

10.1080/10618600.1997.10474727

Cite this

@article{a90db69fd7fe48f8b377da290578bdec,

title = "Bayesian Image Classification with Baddeley{\textquoteright}s Delta Loss",

abstract = "In this article we adopt Baddeley's delta metric as a loss function in Bayesian image restoration and classification. We develop a new algorithm that allows us to approximate the corresponding optimal Bayesian estimator. With this algorithm good practical estimates can be obtained at approximately the same computational cost as traditional estimators like the marginal posterior mode (MPM). A comparison of our proposed classification with MPM shows significant advantages, especially with respect to fine structures.",

keywords = "Asymmetric loss functions, Bayesian inference, Distance between binary images, Image restoration, Markov chain Monte Carlo methods, Metropolis algorithm",

author = "Arnoldo Frigessi and H{\aa}vard Rue",

year = "1997",

month = mar,

doi = "10.1080/10618600.1997.10474727",

language = "English (US)",

volume = "6",

pages = "55--73",

journal = "JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS",

issn = "1061-8600",

publisher = "Taylor and Francis Ltd.",

number = "1",

}

TY - JOUR

T1 - Bayesian Image Classification with Baddeley’s Delta Loss

AU - Frigessi, Arnoldo

AU - Rue, Håvard

PY - 1997/3

Y1 - 1997/3

N2 - In this article we adopt Baddeley's delta metric as a loss function in Bayesian image restoration and classification. We develop a new algorithm that allows us to approximate the corresponding optimal Bayesian estimator. With this algorithm good practical estimates can be obtained at approximately the same computational cost as traditional estimators like the marginal posterior mode (MPM). A comparison of our proposed classification with MPM shows significant advantages, especially with respect to fine structures.

AB - In this article we adopt Baddeley's delta metric as a loss function in Bayesian image restoration and classification. We develop a new algorithm that allows us to approximate the corresponding optimal Bayesian estimator. With this algorithm good practical estimates can be obtained at approximately the same computational cost as traditional estimators like the marginal posterior mode (MPM). A comparison of our proposed classification with MPM shows significant advantages, especially with respect to fine structures.

KW - Asymmetric loss functions

KW - Bayesian inference

KW - Distance between binary images

KW - Image restoration

KW - Markov chain Monte Carlo methods

KW - Metropolis algorithm

UR - http://www.scopus.com/inward/record.url?scp=0031537121&partnerID=8YFLogxK

U2 - 10.1080/10618600.1997.10474727

DO - 10.1080/10618600.1997.10474727

M3 - Article

AN - SCOPUS:0031537121

SN - 1061-8600

VL - 6

SP - 55

EP - 73

JO - JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS

JF - JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS

IS - 1

ER -

Bayesian Image Classification with Baddeley’s Delta Loss

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this