Wavelet based image denoising with a mixture of Gaussian distributions with local parameters

H. Rabbani, M. Vafadoost, Ivan Selesnick

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

Abstract

The performance of various estimators, such as maximum a posteriori (MAP) is strongly dependent on correctness of the proposed model for noise-free data distribution. Therefore, the selection of a proper model for distribution of wavelet coefficients is very important in the wavelet based image denoising. This paper presents a new image denoising algorithm based on the modeling of wavelet coefficients in each subband with a mixture of Gaussian probability density functions (pdfs) that parameters of mixture model are local. The mixture model is able to capture the heavy-tailed nature of wavelet coefficients and the local parameters can model the empirically observed correlation between the coefficient amplitudes. Therefore, by using this relatively new statistical model, we are able to better model statistical property of wavelet coefficients. Within this framework, we describe a novel method for image denoising based on designing a MAP estimator, which relies on the mixture distributions with high local correlation. The simulation results show that our proposed technique achieves better performance than several published methods both visually and in terms of peak signal-to-noise ratio (PSNR).

Original languageEnglish (US)
Title of host publicationProceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications
Pages85-88
Number of pages4
DOIs
StatePublished - 2006
EventELMAR-2006 - 48th International Symposium ELMAR-2006 focused on Multimedia Signal Processing and Communications - Zadar, Croatia
Duration: Jun 7 2006Jun 9 2006

Other

OtherELMAR-2006 - 48th International Symposium ELMAR-2006 focused on Multimedia Signal Processing and Communications
CountryCroatia
CityZadar
Period6/7/066/9/06

Fingerprint

Image denoising
Gaussian distribution
Probability density function
Signal to noise ratio

Keywords

  • Complex Wavelet Transform
  • MAP Estimator
  • Mixture Model

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Rabbani, H., Vafadoost, M., & Selesnick, I. (2006). Wavelet based image denoising with a mixture of Gaussian distributions with local parameters. In Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications (pp. 85-88). [4127494] https://doi.org/10.1109/ELMAR.2006.329521

Wavelet based image denoising with a mixture of Gaussian distributions with local parameters. / Rabbani, H.; Vafadoost, M.; Selesnick, Ivan.

Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications. 2006. p. 85-88 4127494.

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

Rabbani, H, Vafadoost, M & Selesnick, I 2006, Wavelet based image denoising with a mixture of Gaussian distributions with local parameters. in Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications., 4127494, pp. 85-88, ELMAR-2006 - 48th International Symposium ELMAR-2006 focused on Multimedia Signal Processing and Communications, Zadar, Croatia, 6/7/06. https://doi.org/10.1109/ELMAR.2006.329521
Rabbani H, Vafadoost M, Selesnick I. Wavelet based image denoising with a mixture of Gaussian distributions with local parameters. In Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications. 2006. p. 85-88. 4127494 https://doi.org/10.1109/ELMAR.2006.329521
Rabbani, H. ; Vafadoost, M. ; Selesnick, Ivan. / Wavelet based image denoising with a mixture of Gaussian distributions with local parameters. Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications. 2006. pp. 85-88
@inproceedings{506ddec66369425c89bc85586e17ef5e,
title = "Wavelet based image denoising with a mixture of Gaussian distributions with local parameters",
abstract = "The performance of various estimators, such as maximum a posteriori (MAP) is strongly dependent on correctness of the proposed model for noise-free data distribution. Therefore, the selection of a proper model for distribution of wavelet coefficients is very important in the wavelet based image denoising. This paper presents a new image denoising algorithm based on the modeling of wavelet coefficients in each subband with a mixture of Gaussian probability density functions (pdfs) that parameters of mixture model are local. The mixture model is able to capture the heavy-tailed nature of wavelet coefficients and the local parameters can model the empirically observed correlation between the coefficient amplitudes. Therefore, by using this relatively new statistical model, we are able to better model statistical property of wavelet coefficients. Within this framework, we describe a novel method for image denoising based on designing a MAP estimator, which relies on the mixture distributions with high local correlation. The simulation results show that our proposed technique achieves better performance than several published methods both visually and in terms of peak signal-to-noise ratio (PSNR).",
keywords = "Complex Wavelet Transform, MAP Estimator, Mixture Model",
author = "H. Rabbani and M. Vafadoost and Ivan Selesnick",
year = "2006",
doi = "10.1109/ELMAR.2006.329521",
language = "English (US)",
isbn = "9537044033",
pages = "85--88",
booktitle = "Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications",

}

TY - GEN

T1 - Wavelet based image denoising with a mixture of Gaussian distributions with local parameters

AU - Rabbani, H.

AU - Vafadoost, M.

AU - Selesnick, Ivan

PY - 2006

Y1 - 2006

N2 - The performance of various estimators, such as maximum a posteriori (MAP) is strongly dependent on correctness of the proposed model for noise-free data distribution. Therefore, the selection of a proper model for distribution of wavelet coefficients is very important in the wavelet based image denoising. This paper presents a new image denoising algorithm based on the modeling of wavelet coefficients in each subband with a mixture of Gaussian probability density functions (pdfs) that parameters of mixture model are local. The mixture model is able to capture the heavy-tailed nature of wavelet coefficients and the local parameters can model the empirically observed correlation between the coefficient amplitudes. Therefore, by using this relatively new statistical model, we are able to better model statistical property of wavelet coefficients. Within this framework, we describe a novel method for image denoising based on designing a MAP estimator, which relies on the mixture distributions with high local correlation. The simulation results show that our proposed technique achieves better performance than several published methods both visually and in terms of peak signal-to-noise ratio (PSNR).

AB - The performance of various estimators, such as maximum a posteriori (MAP) is strongly dependent on correctness of the proposed model for noise-free data distribution. Therefore, the selection of a proper model for distribution of wavelet coefficients is very important in the wavelet based image denoising. This paper presents a new image denoising algorithm based on the modeling of wavelet coefficients in each subband with a mixture of Gaussian probability density functions (pdfs) that parameters of mixture model are local. The mixture model is able to capture the heavy-tailed nature of wavelet coefficients and the local parameters can model the empirically observed correlation between the coefficient amplitudes. Therefore, by using this relatively new statistical model, we are able to better model statistical property of wavelet coefficients. Within this framework, we describe a novel method for image denoising based on designing a MAP estimator, which relies on the mixture distributions with high local correlation. The simulation results show that our proposed technique achieves better performance than several published methods both visually and in terms of peak signal-to-noise ratio (PSNR).

KW - Complex Wavelet Transform

KW - MAP Estimator

KW - Mixture Model

UR - http://www.scopus.com/inward/record.url?scp=46149110564&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=46149110564&partnerID=8YFLogxK

U2 - 10.1109/ELMAR.2006.329521

DO - 10.1109/ELMAR.2006.329521

M3 - Conference contribution

SN - 9537044033

SN - 9789537044039

SP - 85

EP - 88

BT - Proceedings ELMAR-2006 - 48th International Symposium ELMAR-2006 Focused on Multimedia Signal Processing and Communications

ER -