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 language | English (US) |
---|---|
Pages (from-to) | II/1261-II/1264 |
Journal | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings |
Volume | 2 |
DOIs | |
State | Published - 2002 |
Event | 2002 IEEE International Conference on Acoustic, Speech and Signal Processing - Orlando, FL, United States Duration: May 13 2002 → May 17 2002 |
ASJC Scopus subject areas
- Software
- Signal Processing
- Electrical and Electronic Engineering