Feature-Preserving MRI Denoising :
A Nonparametric Empirical-Bayes Approach
A Nonparametric Empirical-Bayes Approach
Feature-Preserving MRI Denoising: A Nonparametric Empirical Bayes Approach
IEEE Trans. Med. Imaging 2007, 26(9):1242-1255
Bayesian Image Denoising: The Key Problem
(1) How to model the prior (i.e. Markov statistics of noiseless signal) ?
(2) How to estimate the prior ?
Challenges Using Pre-Tuned Parametric Priors
Key Idea : Nonparametric Empirical-Bayes Estimation
(1) Model the prior, i.e. Markov statistics of the noiseless signal, using nonparametric statistical schemes
(2) Estimate the prior from the image that is to be denoised, knowing the noise model
(3) Use this estimate of the prior for optimal Bayesian denoising
Estimating the Prior
Bayesian Denoising by Iterated Conditional Entropy Reduction (ICER)
Denoised Images
Residual Images
= difference between denoised image and noiseless image
= difference between denoised image and noiseless image
Validation : Quantitative (BrainWeb repository)
Denoised a Real MR Image
Related Work
- Suyash P. Awate, Ross T. Whitaker
Unsupervised, Information-Theoretic, Adaptive Image Filtering with Applications to Image Restoration
IEEE Trans. Pattern Analysis & Machine Intelligence (TPAMI) 2006, Vol. 28, Num. 3 , pp. 364-376 - Suyash P. Awate, Ross T. Whitaker
Higher-Order Image Statistics for Unsupervised, Information-Theoretic, Adaptive Image Filtering
IEEE Computer Vision & Pattern Recognition (CVPR), June 2005, vol 2, pp 44-51
Universal discrete denoising: Known channel
IEEE Trans. Information Theory 2005, 51(1):5-28
DUDE and Extensions
The empirical Bayes approach to statistical decision problems
Annals of Mathematical Statistics 1964, 35(1):1-20
An empirical Bayes approach to statistics
In Proc. Third Berkeley Symp. Math. Stat. Prob. 1964, pp.157-164