Shift invariant restoration - An overcomplete maxent MAP framework

P. Ishwar, Pierre Moulin

Research output: Contribution to conferencePaper

Abstract

Translation-invariant denoising was introduced by Coifman and Donoho to overcome Gibbs-type phenomena produced by transform-domain shrinkage estimators in the vicinity of signal discontinuities. Shrinkage estimators are in general not shift-invariant. Shift-invariant denoising consists of a simple averaging of the shrinkage estimates over a family of cyclic spatial-shifts of the image. Shift-invariant denoising is denoising in an overcomplete basis, and work in this area has been devoted towards finding a best basis in the overcomplete family. This paper presents a Maximum A Posteriori (MAP) framework for shift-invariant restoration of images using the maximum-entropy prior consistent with moment constraints on the transform coefficients in different subbands. The simple averaging of estimates in the classical shift-invariant denoising can then be shown to be a certain limiting case within this framework.

Original languageEnglish (US)
StatePublished - Dec 1 2000
EventInternational Conference on Image Processing (ICIP 2000) - Vancouver, BC, Canada
Duration: Sep 10 2000Sep 13 2000

Other

OtherInternational Conference on Image Processing (ICIP 2000)
CountryCanada
CityVancouver, BC
Period9/10/009/13/00

Fingerprint

Restoration
Entropy
Mathematical transformations

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Ishwar, P., & Moulin, P. (2000). Shift invariant restoration - An overcomplete maxent MAP framework. Paper presented at International Conference on Image Processing (ICIP 2000), Vancouver, BC, Canada.

Shift invariant restoration - An overcomplete maxent MAP framework. / Ishwar, P.; Moulin, Pierre.

2000. Paper presented at International Conference on Image Processing (ICIP 2000), Vancouver, BC, Canada.

Research output: Contribution to conferencePaper

Ishwar, P & Moulin, P 2000, 'Shift invariant restoration - An overcomplete maxent MAP framework' Paper presented at International Conference on Image Processing (ICIP 2000), Vancouver, BC, Canada, 9/10/00 - 9/13/00, .
Ishwar P, Moulin P. Shift invariant restoration - An overcomplete maxent MAP framework. 2000. Paper presented at International Conference on Image Processing (ICIP 2000), Vancouver, BC, Canada.
Ishwar, P. ; Moulin, Pierre. / Shift invariant restoration - An overcomplete maxent MAP framework. Paper presented at International Conference on Image Processing (ICIP 2000), Vancouver, BC, Canada.
@conference{7649c56d6b494fd297dd798b1fb6be05,
title = "Shift invariant restoration - An overcomplete maxent MAP framework",
abstract = "Translation-invariant denoising was introduced by Coifman and Donoho to overcome Gibbs-type phenomena produced by transform-domain shrinkage estimators in the vicinity of signal discontinuities. Shrinkage estimators are in general not shift-invariant. Shift-invariant denoising consists of a simple averaging of the shrinkage estimates over a family of cyclic spatial-shifts of the image. Shift-invariant denoising is denoising in an overcomplete basis, and work in this area has been devoted towards finding a best basis in the overcomplete family. This paper presents a Maximum A Posteriori (MAP) framework for shift-invariant restoration of images using the maximum-entropy prior consistent with moment constraints on the transform coefficients in different subbands. The simple averaging of estimates in the classical shift-invariant denoising can then be shown to be a certain limiting case within this framework.",
author = "P. Ishwar and Pierre Moulin",
year = "2000",
month = "12",
day = "1",
language = "English (US)",
note = "International Conference on Image Processing (ICIP 2000) ; Conference date: 10-09-2000 Through 13-09-2000",

}

TY - CONF

T1 - Shift invariant restoration - An overcomplete maxent MAP framework

AU - Ishwar, P.

AU - Moulin, Pierre

PY - 2000/12/1

Y1 - 2000/12/1

N2 - Translation-invariant denoising was introduced by Coifman and Donoho to overcome Gibbs-type phenomena produced by transform-domain shrinkage estimators in the vicinity of signal discontinuities. Shrinkage estimators are in general not shift-invariant. Shift-invariant denoising consists of a simple averaging of the shrinkage estimates over a family of cyclic spatial-shifts of the image. Shift-invariant denoising is denoising in an overcomplete basis, and work in this area has been devoted towards finding a best basis in the overcomplete family. This paper presents a Maximum A Posteriori (MAP) framework for shift-invariant restoration of images using the maximum-entropy prior consistent with moment constraints on the transform coefficients in different subbands. The simple averaging of estimates in the classical shift-invariant denoising can then be shown to be a certain limiting case within this framework.

AB - Translation-invariant denoising was introduced by Coifman and Donoho to overcome Gibbs-type phenomena produced by transform-domain shrinkage estimators in the vicinity of signal discontinuities. Shrinkage estimators are in general not shift-invariant. Shift-invariant denoising consists of a simple averaging of the shrinkage estimates over a family of cyclic spatial-shifts of the image. Shift-invariant denoising is denoising in an overcomplete basis, and work in this area has been devoted towards finding a best basis in the overcomplete family. This paper presents a Maximum A Posteriori (MAP) framework for shift-invariant restoration of images using the maximum-entropy prior consistent with moment constraints on the transform coefficients in different subbands. The simple averaging of estimates in the classical shift-invariant denoising can then be shown to be a certain limiting case within this framework.

UR - http://www.scopus.com/inward/record.url?scp=0034442397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034442397&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0034442397

ER -