Image Denoising Based on A Mixture of Bivariate Gaussian Models in Complex Wavelet Domain

Recently, it has been shown that algorithms exploiting dependencies between coefficients for modeling probability density function (pdf) of wavelet coefficients, could achieve better results for image denoising in wavelet domain compared with the ones based on the independence assumption. In this co...

Full description

Saved in:
Bibliographic Details
Published in2006 3rd IEEE/EMBS International Summer School on Medical Devices and Biosensors pp. 149 - 153
Main Authors Rabbani, H., Vafadoost, M., Selesnick, I., Gazor, S.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.09.2006
Subjects
Bayesian methods
Biomedical engineering
Bivariate pdf
Complex Wavelet Transform
Context modeling
Gaussian noise
Image denoising
Laplace equations
MAP Estimator
Mixture Model
Noise reduction
PSNR
Wavelet coefficients
Wavelet domain
Online AccessGet full text

Cover

More Information
Summary:Recently, it has been shown that algorithms exploiting dependencies between coefficients for modeling probability density function (pdf) of wavelet coefficients, could achieve better results for image denoising in wavelet domain compared with the ones based on the independence assumption. In this context, we design a bivariate maximum a posteriori (MAP) estimator which relies on a mixture of bivariate Gaussian models. This model not only is bivariate but also is mixture and therefore, using this new statistical model, we are able to better capture heavy-tailed natures of the data as well as the interscale dependencies of wavelet coefficients. 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).
ISBN:9780780397866
078039786X
DOI:10.1109/ISSMDBS.2006.360121