@article {176, title = {Quasi-Monte Carlo estimation approach for denoising MRI data based on regional statistics}, journal = {IEEE Transactions on Biomedical Engineering}, volume = {58}, year = {2011}, month = {1076 - 1083}, abstract = {

An important postprocessing step for MR data is noise reduction. Noise in MR data is difficult to suppress due to its signal-dependence. To address this issue, a novel stochastic approach to noise reduction for MR data is presented. The estimation of the noise-free signal is formulated as a general Bayesian least-squares estimation problem and solved using a quasi-Monte Carlo method that takes into account the statistical characteristics of the underlying noise and the regional statistics of the observed signal in a data-adaptive manner. A set of experiments were per formed to compare the proposed quasi-Monte Carlo estimation (QMCE) method to state-of-the-art wavelet-based MR noise reduction (WAVE) and nonlocal means MR noise reduction (NLM) methods using MR data volumes with synthetic noise, as well as real noise-contaminated MR data. Experimental results show that QMCE is capable of achieving state-of-the-art performance when compared to WAVE and NLM methods quantitatively in SNR, mean structural similarity (MSSIM), and contrast measures. Visual comparisons show that QMCE provides effective noise suppression, while better preserving tissue structural boundaries and restoring contrast.

}, keywords = {Bayes methods, biomedical MRI, contrast measures, data adaptive method, general Bayesian least-squares estimation, Image Denoising, mean structural similarity, medical image processing, Monte Carlo methods, MR data post processing step, noise-free signal, nonlocal means MR noise reduction, QMCE, quasi Monte Carlo estimation approach, regional statistics, stochastic approach, stochastic processes, tissue structural boundaries, WAVE, wavelet transforms, wavelet-based MR noise reduction}, issn = {0018-9294}, doi = {http://dx.doi.org/10.1109/TBME.2010.2048325}, author = {A Wong and A Mishra} } @article {177, title = {Quasi-Monte Carlo estimation approach for denoising MRI data based on regional statistics}, journal = {IEEE Transactions on Biomedical Engineering}, volume = {58}, year = {2011}, month = {1076 - 1083}, abstract = {

An important postprocessing step for MR data is noise reduction. Noise in MR data is difficult to suppress due to its signal-dependence. To address this issue, a novel stochastic approach to noise reduction for MR data is presented. The estimation of the noise-free signal is formulated as a general Bayesian least-squares estimation problem and solved using a quasi-Monte Carlo method that takes into account the statistical characteristics of the underlying noise and the regional statistics of the observed signal in a data-adaptive manner. A set of experiments were per formed to compare the proposed quasi-Monte Carlo estimation (QMCE) method to state-of-the-art wavelet-based MR noise reduction (WAVE) and nonlocal means MR noise reduction (NLM) methods using MR data volumes with synthetic noise, as well as real noise-contaminated MR data. Experimental results show that QMCE is capable of achieving state-of-the-art performance when compared to WAVE and NLM methods quantitatively in SNR, mean structural similarity (MSSIM), and contrast measures. Visual comparisons show that QMCE provides effective noise suppression, while better preserving tissue structural boundaries and restoring contrast.

}, keywords = {Bayes methods, biomedical MRI, contrast measures, data adaptive method, general Bayesian least-squares estimation, Image Denoising, mean structural similarity, medical image processing, Monte Carlo methods, MR data post processing step, noise-free signal, nonlocal means MR noise reduction, QMCE, quasi Monte Carlo estimation approach, regional statistics, stochastic approach, stochastic processes, tissue structural boundaries, WAVE, wavelet transforms, wavelet-based MR noise reduction}, issn = {0018-9294}, doi = {http://dx.doi.org/10.1109/TBME.2010.2048325}, author = {A Wong and A Mishra} } @inproceedings {291, title = {Video pause detection using wavelets}, booktitle = {6th Canadian Conference on Computer and Robot Vision}, year = {2009}, month = {05/2009}, address = {Kelowna, British Columbia, Canada}, abstract = {

As the volume of digital video captured and stored continues to increase, research efforts have focused on content management systems for video indexing and retrieval applications. A first step in generic video processing is shot boundary detection. This paper addresses a novel algorithm for abrupt shot (cut/pause) detection-especially on frames with similar statistics-based on the wavelet transform and content entropy. The algorithm has been successfully tested on some video categories including sport and live videos. Its quantitative performance has been compared to other known methods including pixel, histogram, frequency domain and statistics difference. In each test, the proposed wavelet method outperforms the others.

}, keywords = {content entropy, content management, content management system, digital video capture, entropy, indexing, shot boundary detection, statistical analysis, statistics-based wavelet transform, video indexing, Video Recording, video retrieval, wavelet transforms}, doi = {http://dx.doi.org/10.1109/CRV.2009.20}, author = {S Zaboli and D A. Clausi} } @inproceedings {327, title = {Automatic registration of inter-band and inter-sensor images using robust complex wavelet feature representations}, booktitle = {5th Workshop on Pattern Recognition for Remote Sensing}, year = {2008}, month = {12/2008}, address = {Tampa, Florida, USA}, abstract = {

A robust method for registering inter-band and inter-sensor remote sensing images has been designed and implemented. The proposed method introduces noise-resilient and contrast invariant control point detection and control point matching schemes based on robust complex wavelet feature representations. Furthermore, an iterative refinement scheme is introduced for achieving improved control point pair localization and mapping function estimation between the images being registered. The registration accuracy of the proposed method was demonstrated on the registration of multi-spectral optical and synthetic aperture radar (SAR) images. The proposed method achieves better registration accuracy when compared with the state-of-the-art MSSD and ARRSI registration algorithms.

}, keywords = {ARRSI, automatic registration, contrast invariant control point detection, control point matching schemes, control point pair localization, estimation theory, Feature Extraction, image matching, image registration, image representation, images registration, inter-band images, inter-sensor images, iterative methods, iterative refinement scheme, mapping function estimation, MSSD, multispectral optical images, registration algorithms, Remote Sensing, remote sensing images, robust complex wavelet feature representations, synthetic aperture radar, synthetic aperture radar images, wavelet transforms}, doi = {http://dx.doi.org/10.1109/PRRS.2008.4783164}, author = {A Wong and D A. Clausi and P Fieguth} } @inproceedings {343, title = {Correlated non-linear wavelet shrinkage}, booktitle = {15th IEEE International Conference on Image Processing, 2008}, year = {2008}, abstract = {

This paper examines non-linear shrinkage methods specifically taking into account the correlation structure of the multiresolution wavelet coefficients. In contrast to hidden Markov trees, which model the relationship of wavelet variance from scale to scale, here we wish to take advantage of coefficient correlation. A linear shrinkage based on the LLS (Linear Least Square) estimator, employing a sample correlation scheme, is tested and verified to have an aesthetic denoising performance. Then, state-of-the-art independent shrinkage functions are applied to exploit the efficiency of such techniques and to introduce non-linearity into the algorithm to compensate for non-Gaussianity of the wavelet statistics. The performance of the non-linear shrinkage technique, as used individually and together with the linear correlated approach, are illustrated.

}, keywords = {coefficient correlation, correlated non-linear wavelet shrinkage, hidden Markov models, Image Denoising, linear least square estimator, multiresolution wavelet coefficients, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2008.4712263}, author = {M Amiri and Z Azimifar and P Fieguth} } @inproceedings {321, title = {A perceptually adaptive approach to image denoising using anisotropic non-local means}, booktitle = {15th IEEE International Conference on Image Processing}, year = {2008}, month = {10/2008}, address = {San Diego, California, USA}, abstract = {

This paper introduces a novel perceptually adaptive approach to image denoising using anisotropic non-local means. In the classical non-local means image denoising approach, the value of a pixel is determined based on the weighted average of other pixels, where the weights are determined based on a fixed isotropically weighted similarity function between the local neighborhoods. In the proposed algorithm, we demonstrate that noticeably improved perceptual quality can be achieved through the use of adaptive anisotropically weighted similarity functions between local neighborhoods. This is accomplished by adapting the similarity weighing function in an anisotropic manner based on the perceptual characteristics of the underlying image content derived efficiently based on the Mexican Hat wavelet. Experimental results show that the proposed method can be used to provide improved perceptual quality in the denoised image both quantitatively and qualitatively when compared to existing methods.

}, keywords = {adaptive anisotropically weighted similarity function, anisotropic nonlocal mean approach, Image Denoising, Mexican Hat wavelet transform, perceptual adaptive approach, perceptual quality, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2008.4711810}, author = {A Wong and D A. Clausi and P Fieguth} } @inproceedings {395, title = {A combined Bayesshrink wavelet-ridgelet technique for image denoising}, booktitle = {2006 IEEE International Conference on Multimedia and Expo}, year = {2006}, abstract = {

In this paper a combined Bayesshrink wavelet-ridgelet de-noising method is presented. In our previous work we have showed that Bayesshrink ridgelet performs better than Visushrink ridgelet and Visushrink wavelet. Although our Bayesshrink ridgelet technique performs somewhat poorer in comparison with Bayesshrink wavelet, based on SNR, visually it produces smoother results, especially for images with straight lines. In the proposed method Bayesshrink wavelet is combined with Bayesshrink ridgelet denoising method which performs better than each filter individually. The proposed combined denoising method gains the advantage of each filter in its specific domain, i.e., wavelet for natural and ridgelet for straight regions, and produces better and smoother results, both visually and in terms of SNR

}, keywords = {Bayes methods, Bayesshrink wavelet ridgelet technique, filtering theory, Image Denoising, image denoising method, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICME.2006.262931}, author = {N Nezamoddin Kachouie and P Fieguth} } @inproceedings {420, title = {Correlated wavelet shrinkage: models of local random fields across multiple resolutions}, booktitle = {International Conference on Image Analysis and Recognition}, year = {2005}, abstract = {

This paper proposes a novel correlated shrinkage method based on wavelet joint statistics. Our objective is to demonstrate effectiveness of the wavelet correlation models [Z. Azimifar et al., 2004] in estimating the original signal from a noising observation. Simulation results are given to show the advantage of the new correlated shrinkage function. In comparison with the popular nonlinear shrinkage algorithms, it improves the denoised results.

}, keywords = {correlated wavelet shrinkage, Image Denoising, image resolution, image resolutions, local random fields, nonlinear shrinkage algorithms, statistical analysis, wavelet correlation models, wavelet joint statistics, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2005.1530352}, author = {Z Azimifar and P Fieguth and E Jernigan} } @inproceedings {421, title = {A Gabor based technique for image denoising}, booktitle = {18th Canadian Conference on Electrical and Computer Engineering}, year = {2005}, address = {Saskatoon}, abstract = {

As an alternative to the wavelet, Gabor function has been used as an efficient representation of two dimensional signals. We are interested in BayesShrink techniques for image denoising, and have shown in our previous work that BayesShrink Ridgelet performs better than VisuShrink ridgelet and VisuShrink wavelet. In this paper, a dyadic Gabor filter bank is combined with BayesShrink method for image denoising. In the proposed method, the noisy image is decomposed to different channels in several levels by a dyadic Gabor filter bank. To recover the image, the corrupting noise is removed by applying the proposed BayesShrink method on the noisy Gabor coefficients. The noise variance is estimated in Gabor domain and the estimated noise is then used to dynamically calculate an individual threshold for each spatio-frequency channel. Finally denoised coefficients are transformed back to reconstruct the image

}, keywords = {Bayes methods, BayesShrink ridgelet, channel bank filters, dyadic Gabor filter banks, Gabor based technique, Gabor filters, Image Denoising, image reconstruction, image representation, noise variance, spatiofrequency channels, two-dimensional signal representation, VisuShrink ridgelet, wavelet transforms}, doi = {http://dx.doi.org/10.1109/CCECE.2005.1557140}, author = {N Nexamoddin Kachouie and P Fieguth} } @inproceedings {663, title = {Adaptive Wiener filtering of noisy images and image sequences}, booktitle = {IEEE International Conference on Image Processing}, year = {2003}, address = {Spain}, abstract = {

In this work, we consider the adaptive Wiener filtering of noisy images and image sequences. We begin by using an adaptive weighted averaging (AWA) approach to estimate the second-order statistics required by the Wiener filter. Experimentally, the resulting Wiener filter is improved by about 1 dB in the sense of peak-to-peak SNR (PSNR). Also, the subjective improvement is significant in that the annoying boundary noise, common with the traditional Wiener filter, has been greatly suppressed. The second, and more substantial, part of this paper extends the AWA concept to the wavelet domain. The proposed AWA wavelet Wiener filter is superior to the traditional wavelet Wiener filter by about 0.5 dB (PSNR). Furthermore, an interesting method to effectively combine the denoising results from both wavelet and spatial domains is shown and discussed. Our experimental results outperform or are comparable to state-of-art methods.

}, keywords = {adaptive filters, adaptive weighted averaging, adaptive Wiener filtering, Image Denoising, image sequences, noisy images, peak-to-peak SNR, second-order statistics, statistical analysis, wavelet domain, wavelet transforms, Wiener filters}, doi = {http://dx.doi.org/10.1109/ICIP.2003.1247253}, author = {F Jin and P Fieguth and L Winger and E Jernigan} } @inproceedings {757, title = {Hierarchical Markov models for wavelet-domain statistics}, booktitle = {2003 IEEE Workshop on Statistical Signal Processing}, year = {2003}, abstract = {

There is a growing realization that modeling wavelet coefficients as statistically independent may be a poor assumption. Thus, this paper investigates two efficient models for wavelet coefficient coupling. Spatial statistics which are Markov (commonly used for textures and other random imagery) do not preserve their Markov properties in the wavelet domain; that is, the wavelet-domain covariance Pw does not have a sparse inverse. The main theme of this work is to investigate the approximation of Pw by hierarchical Markov and non-Markov models.

}, keywords = {correlation methods, hierarchical Markov models, image processing, Markov processes, spatial statistics, statistical analysis, wavelet coefficient coupling, wavelet transforms, wavelet-domain covariance, wavelet-domain statistics}, doi = {http://dx.doi.org/10.1109/SSP.2003.1289393}, author = {Z Azimifar and P Fieguth and E Jernigan} } @inproceedings {594, title = {Data fusion of SSM/I channels using multiresolution wavelet transform}, booktitle = {IEEE International Geoscience And Remote Sensing Symposium}, year = {2002}, abstract = {

This paper presents an approach to the fusion of SSM/I (Special Sensor Microwave/Imager) data from different resolutions, based on the prior statistical information about the data. The result is an estimated field that lives in a finer scale than any of the measurements. We apply a Wavelet Transform that increases speed and decreases memory requirements by sparsifying and preconditioning the statistics. This approach makes feasible the use of reprogrammable FPGA implementations for onboard satellite data processing, which greatly enhances flexibility and, most importantly, reduces communication burdens by limiting the extent to which raw, unprocessed data are transmitted to the ground.

}, keywords = {19 to 85.5 GHz, atmosphere, atmospheric techniques, EHF, geophysical measurement technique, geophysical signal processing, geophysical techniques, image fusion, land surface, meteorology, microwave radiometry, multiresolution wavelet transform, ocean, oceanographic techniques, preconditioning, prior statistical information, satellite remote sensing, sea surface, sensor fusion, SHF, sparsifying, Special Sensor Microwave/Imager, SSM/I, statistics, terrain mapping, wavelet transforms}, doi = {http://dx.doi.org/10.1109/IGARSS.2002.1026564}, author = {V K. Mehta and C M. Hammock and P Fieguth and H Krim} } @inproceedings {598, title = {Towards random field modeling of wavelet statistics}, booktitle = {International Conference on Image Processing}, year = {2002}, address = {Rochester, NY}, abstract = {

The paper investigates the statistical characterization of signals and images in the wavelet domain. In particular, in contrast to common decorrelated-coefficient models, we find that the correlation between wavelet scales can be surprisingly substantial, even across several scales. We investigate possible choices of statistical-interaction models. One efficient and fast strategy which describes the wavelet-based statistical correlations is illustrated. Finally, the effectiveness of the proposed tool towards an efficient hierarchical MRF (Markov random field) modeling of within-scale neighborhoods and across-scale dependencies is demonstrated.

}, keywords = {image characterization, image processing, Markov processes, Markov random field modeling, MRF, signal characterization, statistical analysis, statistical image processing, wavelet statistics, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2002.1038035}, author = {Z Azimifar and P Fieguth and E Jernigan} } @inproceedings {533, title = {Modeling the correlation structure of images in the wavelet domain}, booktitle = {14th Canadian Conference on Electrical and Computer Engineering}, year = {2001}, address = {Toronto}, abstract = {

In this paper we investigate the correlation structure of the wavelet coefficients corresponding to random fields. The context of this work is the study of Bayesian approaches to wavelet shrinkage for the purposes of image denoising. This paper concentrates on both within-scale and across-scale statistical dependencies for a variety of wavelets and random fields, with examples provided for both 1-D and 2-D signals. The results show the whitening effect of the wavelet transform to be quite clear-even for particular highly correlated spatial processes the within-scale correlation decays exponentially fast, however the correlation between scales is surprisingly substantial, even for separations several scales apart. Our goal, initiated in this paper, is the development of an efficient random field model, describing these statistical correlations, and the demonstration of its effectiveness in the context of Bayesian wavelet shrinkage for signal and image denoising

}, keywords = {1-D signals, 2-D signals, across-scale statistical dependencies, Bayes methods, Bayesian approaches, correlation structure, correlation theory, Image Denoising, image enhancement, random field model, random fields, wavelet coefficients, wavelet domain, wavelet shrinkage, wavelet transforms, white noise, whitening effect, within-scale statistical dependencies}, doi = {http://dx.doi.org/10.1109/CCECE.2001.933599}, author = {Z Azimifar and P Fieguth and E Jernigan} } @inproceedings {568, title = {Wavelet shrinkage with correlated wavelet coefficients}, booktitle = {International Conference on Image Processing}, year = {2001}, address = {Greece}, abstract = {

This paper investigates the statistical characterization of multiscale wavelet coefficients corresponding to random signals and images. Virtually all approaches to wavelet shrinkage model the wavelet coefficients as independent; we challenge that assumption and demonstrate several cases where substantial correlations may be present in the wavelet domain. In particular, the correlation between scales can be surprisingly substantial, even for pixels separated by several scales. Our goal is to develop an efficient random field model describing these statistical correlations, and demonstrate its effectiveness in the context of Bayesian wavelet shrinkage for signal and image denoising

}, keywords = {Bayes methods, Bayesian wavelet shrinkage, correlated wavelet coefficients, Image Denoising, image resolution, multiscale wavelet coefficients, random field model, signal denoising, smoothing methods, statistical analysis, statistical correlations, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2001.958076}, author = {Z Azimifar and P Fieguth and E Jernigan} } @inproceedings {501, title = {Fast retrieval methods for images with significant variations}, booktitle = {International Conference on Image Processing}, year = {2000}, abstract = {

Fast image retrieval is the key to success for operations on large image databases, and a great many techniques have been developed for efficient retrieval. However, most of these methods are tailored to visual scenes or to images having limited variations. We investigate the searching of enormous databases (of up to 10 7 images) for the matching and identification of precious stones (principally diamonds). Because of the size of the database, we propose a hierarchy of classifiers, which successively prune candidate images such that the more complex classifiers are required to test only tiny portions of the data. The new classifier developed here applies a wavelet transform to image histograms and is capable of rejecting 99.9\% of bad matches.

}, keywords = {classifiers hierarchy, database searching, diamond, diamonds, fast image retrieval methods, image classification, image histograms, image matching, image retrieval, large image databases, precious stones identification, precious stones matching, visual databases, visual scenes, wavelet transform, wavelet transforms}, doi = {http://dx.doi.org/10.1109/ICIP.2000.899470}, author = {P Fieguth and R Wan} }