A bivariate shrinkage function for wavelet-based denoising

Levent Şendur, Ivan Selesnick

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Most simple nonlinear thresholding rules for wavelet-based denoising assume the wavelet coefficients are independent. However, wavelet coefficients of natural images have significant dependency. In this paper, a new heavy-tailed bivariate pdf is proposed to model the statistics of wavelet coefficients, and a simple nonlinear threshold function (shrinkage function) is derived from the pdf using Bayesian estimation theory. The new shrinkage function does not assume the independence of wavelet coefficients.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2
StatePublished - 2002
Event2002 IEEE International Conference on Acoustic, Speech and Signal Processing - Orlando, FL, United States
Duration: May 13 2002May 17 2002

Other

Other2002 IEEE International Conference on Acoustic, Speech and Signal Processing
CountryUnited States
CityOrlando, FL
Period5/13/025/17/02

Fingerprint

shrinkage
coefficients
Statistics
statistics
thresholds

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Şendur, L., & Selesnick, I. (2002). A bivariate shrinkage function for wavelet-based denoising. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. 2)

A bivariate shrinkage function for wavelet-based denoising. / Şendur, Levent; Selesnick, Ivan.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 2002.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Şendur, L & Selesnick, I 2002, A bivariate shrinkage function for wavelet-based denoising. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. 2, 2002 IEEE International Conference on Acoustic, Speech and Signal Processing, Orlando, FL, United States, 5/13/02.
Şendur L, Selesnick I. A bivariate shrinkage function for wavelet-based denoising. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2. 2002
Şendur, Levent ; Selesnick, Ivan. / A bivariate shrinkage function for wavelet-based denoising. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. 2 2002.
@inproceedings{430dd4d44793427a85ee8c0e5cf77773,
title = "A bivariate shrinkage function for wavelet-based denoising",
abstract = "Most simple nonlinear thresholding rules for wavelet-based denoising assume the wavelet coefficients are independent. However, wavelet coefficients of natural images have significant dependency. In this paper, a new heavy-tailed bivariate pdf is proposed to model the statistics of wavelet coefficients, and a simple nonlinear threshold function (shrinkage function) is derived from the pdf using Bayesian estimation theory. The new shrinkage function does not assume the independence of wavelet coefficients.",
author = "Levent Şendur and Ivan Selesnick",
year = "2002",
language = "English (US)",
volume = "2",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - A bivariate shrinkage function for wavelet-based denoising

AU - Şendur, Levent

AU - Selesnick, Ivan

PY - 2002

Y1 - 2002

N2 - Most simple nonlinear thresholding rules for wavelet-based denoising assume the wavelet coefficients are independent. However, wavelet coefficients of natural images have significant dependency. In this paper, a new heavy-tailed bivariate pdf is proposed to model the statistics of wavelet coefficients, and a simple nonlinear threshold function (shrinkage function) is derived from the pdf using Bayesian estimation theory. The new shrinkage function does not assume the independence of wavelet coefficients.

AB - Most simple nonlinear thresholding rules for wavelet-based denoising assume the wavelet coefficients are independent. However, wavelet coefficients of natural images have significant dependency. In this paper, a new heavy-tailed bivariate pdf is proposed to model the statistics of wavelet coefficients, and a simple nonlinear threshold function (shrinkage function) is derived from the pdf using Bayesian estimation theory. The new shrinkage function does not assume the independence of wavelet coefficients.

UR - http://www.scopus.com/inward/record.url?scp=0036298783&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036298783&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0036298783

VL - 2

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -