Wavelet-domain hidden Markov models have proven to be useful tools for statistical signal and image processing. The hidden Markov tree (HMT) model captures the key features of the joint density of the wavelet coefficients of real-world data. One potential drawback to the HMT framework is the need for computationally expensive iterative training (using the ExpectationMaximization algorithm, for example). In this paper, we propose two reduced-parameter HMT models that capture the general structure of a broad class of real-world images. In the image HMT (iHMT) model we use the fact that for a large class of images the structure of the HMT is self-similar across scale. This allows us to reduce the complexity of the iHMT to just nine easily trained parameters (independent of the size of the image and the number of wavelet scales). In the universal HMT (uHMT) we take a Bayesian approach and fix these nine parameters. The uHMT requires no training of any kind. While simple, we show using a series of image estimation/denoising experiments that these two new models retain nearly all of the key structure modeled by the full HMT. Finally, we propose a fast shift-invariant HMT estimation algorithm that outperforms all other wavelet-based estimators in the current literature, both in mean-square error and visual metrics.
|
2354
|
Stochastic relaxation, Gibbs distribution and the Bayesian restoration of images
– Geman, Geman
- 1984
|
|
2151
|
A tutorial on hidden markov models and selected apllications in speech recognition
– Rabiner
- 1990
|
|
1554
|
A theory for multiresolution signal decomposition: the wavelet representation
– Mallat
- 1989
|
|
1115
|
lectures on wavelets
– Ten
- 1992
|
|
952
|
A wavelet tour of signal processing
– Mallat
- 2001
|
|
950
|
Embedded image coding using zerotrees of wavelet coefficients
– Shapiro
- 1993
|
|
377
|
Mixture densities, maximum likelihood, and the EM algorithm
– Redner, Walker
- 1984
|
|
358
|
Characterization of signals from multiscale edges
– Mallat, Zhong
- 1992
|
|
339
|
Wavelets and subband coding
– Vetterli, Kovacevic
- 1995
|
|
230
|
Image compression through wavelet transform coding
– DeVore, Jawerth, et al.
- 1992
|
|
200
|
Wavelet-based statistical signal processing using Hidden Markov Models
– Crouse, Nowak, et al.
- 1998
|
|
197
|
Wavelets and Operators
– Meyer
- 1993
|
|
171
|
Graphical Models for Machine Learning and Digital Communication
– Frey
- 1998
|
|
157
|
Translation4nvariant denoising
– Coifman, Donoho
- 1995
|
|
156
|
Introduction to Wavelets and Wavelet Transforms: A Primer
– Burrus, Gopinath, et al.
- 1998
|
|
140
|
Wavelet thresholding via a Bayesian approach
– Abramovich, Sapatinas, et al.
- 1998
|
|
122
|
Analysis of multiresolution image denoising schemes using generalized Gaussian and complexity priors
– Moulin, Liu
- 1999
|
|
118
|
Adaptive Bayesian wavelet shrinkage
– Chipman, Kolaczyk, et al.
- 1997
|
|
106
|
Unconditional bases are optimal bases for data compression and for statistical estimation
– Donoho
- 1993
|
|
99
|
Nonlinear Wavelet Image Processing: Variational Problems, Compression, and Noise Removal through Wavelet Shrinkage
– Chambolle, DeVore, et al.
- 1998
|
|
89
|
Statistical models for images: Compression, restoration and synthesis
– Simoncelli
- 1997
|
|
82
|
Low-complexity image denoising based on statistical modeling of wavelet coefficients
– Kivanc, Kozintsev, et al.
- 1999
|
|
64
|
Noise reduction using an undecimated discrete wavelet transform
– Lang, Guo, et al.
- 1996
|
|
42
|
Multiscale image segmentation using wavelet-domain hidden markov models," submitted to
– Choi, Baraniuk
- 1999
|
|
30
|
Nonlinear processing of a shift invariant dwt for noise reduction
– Lang, Guo, et al.
- 1995
|
|
26
|
Translation-invariant de-noising,” in Wavelets and Statistics
– Coifman, Donoho
- 1995
|
|
17
|
Nonlinear approximation of random functions
– Cohen, D’Ales
- 1997
|
|
12
|
Statistics of natural images: Scaling in the woods,” Phys
– Ruderman, Bialek
- 1994
|
|
11
|
Which stochastic models allow Baum-Welch training
– Lucke
- 1996
|
|
10
|
Parameter estimation of dependence tree models using the EM algorithm
– Ronen, Rohlicek, et al.
- 1995
|
|
5
|
Stochastic expansions in and overcomplete wavelet dictionary,” Probability Theory and Related Fields
– Abramovich, Sapatinas, et al.
- 2000
|
|
4
|
Bayesian wavelet-based signal estimation using noninformative priors
– Figueiredo, Nowak
- 1998
|
|
4
|
Wavelet-domain statistical models and Besov spaces
– Choi, Baraniuk
- 1999
|
|
1
|
On lacunary wavelet series," preprint
– Jaffard
|
|
1
|
Tree-structured lacunary wavelet series," preprint
– Baraniuk
|
|
1
|
images [Online]. Available: www.dsp.rice.edu/software/WHMT
– Test
|
|
1
|
Statistical signal restoration with Ia� wavelet domain prior models
– Dufour, Miller
- 1998
|
|
1
|
A universal hidden Markov tree image model
– Romberg
- 1999
|
|
1
|
Statistical Wavelet Models and Function Spaces, preprint
– Choi, Baraniuk
- 2000
|
|
