Modeling statistical properties of wavelets using a mixture of bivariate cauchy models and its application for image denoising in complex wavelet domain

Hossein Rabbani, Mansur Vafadust, Ivan Selesnick

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

Abstract

In this paper, we design a bivariate maximum a posteriori (MAP) estimator that supposes the prior of wavelet coefficients as a mixture of bivariate Cauchy distributions. This model not only is a mixture but is also bivariate. Since mixture models are able to capture the heavy-tailed property of wavelets and bivaraite distributions can model the intrascale dependences of wavelet coefficients, this bivariate mixture probability density function (pdf) can better capture statistical properties of wavelet coefficients. The simulation results show that our proposed technique achieves better performance than other methods employing non mixture pdfs such as bivariate Cauchy pdf and circular symmetric Laplacian pdf visually and in terms of peak signal-to-noise ratio (PSNR). We also compare our algorithm with several recently published denoising methods and see that it is among the best reported in the literature.

Original languageEnglish (US)
Title of host publicationWavelets XII
Volume6701
DOIs
StatePublished - 2007
EventWavelets XII - San Diego, CA, United States
Duration: Aug 26 2007Aug 29 2007

Other

OtherWavelets XII
CountryUnited States
CitySan Diego, CA
Period8/26/078/29/07

Fingerprint

Image denoising
probability density functions
Probability density function
coefficients
estimators
Signal to noise ratio
signal to noise ratios
simulation

Keywords

  • Bivariate cauchy distribution
  • Complex wavelet transform
  • MAP estimator
  • Mixture model

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Modeling statistical properties of wavelets using a mixture of bivariate cauchy models and its application for image denoising in complex wavelet domain. / Rabbani, Hossein; Vafadust, Mansur; Selesnick, Ivan.

Wavelets XII. Vol. 6701 2007. 67012I.

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

@inproceedings{e4e76608abdb413da07cee690accbd3c,
title = "Modeling statistical properties of wavelets using a mixture of bivariate cauchy models and its application for image denoising in complex wavelet domain",
abstract = "In this paper, we design a bivariate maximum a posteriori (MAP) estimator that supposes the prior of wavelet coefficients as a mixture of bivariate Cauchy distributions. This model not only is a mixture but is also bivariate. Since mixture models are able to capture the heavy-tailed property of wavelets and bivaraite distributions can model the intrascale dependences of wavelet coefficients, this bivariate mixture probability density function (pdf) can better capture statistical properties of wavelet coefficients. The simulation results show that our proposed technique achieves better performance than other methods employing non mixture pdfs such as bivariate Cauchy pdf and circular symmetric Laplacian pdf visually and in terms of peak signal-to-noise ratio (PSNR). We also compare our algorithm with several recently published denoising methods and see that it is among the best reported in the literature.",
keywords = "Bivariate cauchy distribution, Complex wavelet transform, MAP estimator, Mixture model",
author = "Hossein Rabbani and Mansur Vafadust and Ivan Selesnick",
year = "2007",
doi = "10.1117/12.739253",
language = "English (US)",
isbn = "9780819468499",
volume = "6701",
booktitle = "Wavelets XII",

}

TY - GEN

T1 - Modeling statistical properties of wavelets using a mixture of bivariate cauchy models and its application for image denoising in complex wavelet domain

AU - Rabbani, Hossein

AU - Vafadust, Mansur

AU - Selesnick, Ivan

PY - 2007

Y1 - 2007

N2 - In this paper, we design a bivariate maximum a posteriori (MAP) estimator that supposes the prior of wavelet coefficients as a mixture of bivariate Cauchy distributions. This model not only is a mixture but is also bivariate. Since mixture models are able to capture the heavy-tailed property of wavelets and bivaraite distributions can model the intrascale dependences of wavelet coefficients, this bivariate mixture probability density function (pdf) can better capture statistical properties of wavelet coefficients. The simulation results show that our proposed technique achieves better performance than other methods employing non mixture pdfs such as bivariate Cauchy pdf and circular symmetric Laplacian pdf visually and in terms of peak signal-to-noise ratio (PSNR). We also compare our algorithm with several recently published denoising methods and see that it is among the best reported in the literature.

AB - In this paper, we design a bivariate maximum a posteriori (MAP) estimator that supposes the prior of wavelet coefficients as a mixture of bivariate Cauchy distributions. This model not only is a mixture but is also bivariate. Since mixture models are able to capture the heavy-tailed property of wavelets and bivaraite distributions can model the intrascale dependences of wavelet coefficients, this bivariate mixture probability density function (pdf) can better capture statistical properties of wavelet coefficients. The simulation results show that our proposed technique achieves better performance than other methods employing non mixture pdfs such as bivariate Cauchy pdf and circular symmetric Laplacian pdf visually and in terms of peak signal-to-noise ratio (PSNR). We also compare our algorithm with several recently published denoising methods and see that it is among the best reported in the literature.

KW - Bivariate cauchy distribution

KW - Complex wavelet transform

KW - MAP estimator

KW - Mixture model

UR - http://www.scopus.com/inward/record.url?scp=42149120377&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42149120377&partnerID=8YFLogxK

U2 - 10.1117/12.739253

DO - 10.1117/12.739253

M3 - Conference contribution

SN - 9780819468499

VL - 6701

BT - Wavelets XII

ER -