Title: Hidden Markov Tree Model of the Uniform Discrete Curvelet Transform Image for Denoising
1Hidden Markov Tree Model of the Uniform Discrete
Curvelet Transform Image for Denoising
2Introduction
- Curvelet Transform (CandesDonoho 1999)
- Implementation
- Fast Discrete Curvelet Transform (FDCT) (Candes
et. al 2005) in frequency domain - Contourlet (DoVetterli 2005) in time domain
with wavelet-like tree structure - Uniform Discrete Curvelet Transform (UDCT)
(NguyenChauris 2008) in frequency domain with
wavelet-like tree structure
3Implementation
4UDCT Marginal Statistics
Kurtosis 23.71
Kurtosis 24.42
Kurtosis E(x-µ)4/s4 . Kurtosis of Gaussian
3
5Conditional Distribution (1)
- On parent (same position in next level)
P(XPX)
Bow-tie shape? uncorrelated but dependent
6Conditional Distribution (2)
- On parent
- P(XPXpx)
- Kurtosis3.51
- Gaussian
7Hidden Markov Tree (HMT) Model
- Conditional distribution is Gaussian
- X depends on PX
- ? Use HMT to model the coefficients
- HMT model links between the hidden state
variables of parent and children - HMT parameters (parameters of the density
function) can be trained using the
expectation-minimization (EM) algorithm
8Tree Structure of UDCT
9HMT (1)
- c(j,k,n) coefficient in scale j, direction k,
position n - S(j,k,n) hidden state taking on values
- m S or L with density function
P(S(j,k,n)) - Conditioned on S(j,k,n)m, c(j,k,n) is Gaussian
with mean µm(j,k,n) and variance - s2m(j,k,n) (mS?small variance, mL?large
variance)
10HMT (2)
- The total pdf
- P(S(j,k,n)), µm(j,k,n), s2m(j,k,n) can be trained
from the EM algorithm (Crouse et al 1998). - Define T set of P(S(j,k,n)), µm(j,k,n),
s2m(j,k,n)
11Denoising (1)
- Problem formulation y xw
- y?noisy coefficients
- x?denoised coefficients
- w?noise coefficients with known variance
- Want to estimate x from the knowledge of y and
variance of w
12Denoising (2)
- Obtain T from EM algorithm
- The variance of denoised coefficients is
13Denoising (3)
14Denoising Results (1)
PSNR Peak Signal to Noise Ratio
15Denoising Results (2)
SSIM Structure Similarity Index (Wang et. al
2004)
16Denoising Results (3)
Original Noisy (14.14dB)
Wavelet (25.73dB)
(SSIM 0.112)
(SSIM 0.561)
Contourlet (25.85dB) DT-CWT (26.54dB) UDCT
(27.32dB)
(SSIM 0.590) (SSIM 0.579)
(SSIM 0.676)
17Denoising Results (4)
Original Noisy (14.14dB)
Wavelet (23.38dB)
(SSIM
0.184) (SSIM 0.508)
Contourlet (22.94dB) DT-CWT (24.15dB) UDCT
(24.35dB)
(SSIM 0.479) (SSIM 0.557)
(SSIM 0.570)
18Denoising Results (5)
Original Noisy (14.14dB)
Wavelet (25.25dB)
(SSIM
0.110) (SSIM 0.539)
Contourlet (25.51dB) DT-CWT (25.99dB) UDCT
(26.51dB)
(SSIM 0.555) (SSIM 0.553)
(SSIM 0.627)