Browsing ECE Publications by Subject "Image Processing and Pattern analysis"
Now showing items 137 of 37

The 2nuSVM: A CostSensitive Extension of the nuSVM
(20051201)Standard classification algorithms aim to minimize the probability of making an incorrect classification. In many important applications, however, some kinds of errors are more important than others. In this report we ... 
Adaptive Representation of JPEG 2000 Images Using Headerbased Processing
(20020920)To bridge the mismatch between the sizes of images and display devices, we present an efficient and automatic algorithm to create an adaptive image representation called SmartNail. Given a digital image and rectangular ... 
Analysis of Multiscale Texture Segmentation using WaveletDomain Hidden Markov Trees
(19991001)This paper describes a technique for estimating the KullbackLeibler (KL) distance between two Hidden Markov Models (HMMs), and for measuring the quality of the estimator. It also provides some results based on applying ... 
Bayesian TreeStructured Image Modeling using Waveletdomain Hidden Markov Models
(20010701)Waveletdomain 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 probability density of the wavelet ... 
Coding Theoretic Approach to Image Segmentation
(20011020)In this paper, using a coding theoretic approach, we implement Rissanen's concept of minimum description length (MDL) for segmenting an image into piecewise homogeneous regions. Our image model is a Gaussian random field ... 
Coding Theoretic Approach to Image Segmentation
(20010520)Using a coding theoretic approach, we achieve unsupervised image segmentation by implementing Rissanen's concept of Minimum Description Length for estimating piecewise homogeneous regions in images. MDL offers a mathematical ... 
Coherent Image Processing using Quaternion Wavelets
(20050801)We develop a quaternion wavelet transform (QWT) as a new multiscale analysis tool for geometric image features. The QWT is a near shiftinvariant, tight frame representation whose coefficients sport a magnitude and three ... 
Compression Color Space Estimation of JPEG Images using Lattice Basis Reduction
(20011001)Given a color image that was previously JPEGcompressed in some hidden color space, we aim to estimate this unknown compression color space from the image. This knowledge is potentially useful for color image enhancement ... 
Controlling False Alarms with Support Vector Machines
(20060501)We study the problem of designing support vector classifiers with respect to a NeymanPearson criterion. Specifically, given a userspecified level alpha, 0 < alpha < 1, how can we ensure a false alarm rate no greater than ... 
Distributed Image Compression for Sensor Networks using Correspondence Analysis and SuperResolution
(20030901)We outline a distributed coding technique for images captured from sensors with overlapping fields of view in a sensor network. First, images from correlated views are roughly registered (relative to a sensor of primary ... 
EstimationQuantization Geometry Coding Using Normal Meshes
(20030301)We propose a new algorithm for compressing threedimensional triangular mesh data used for representing surfaces. We apply the EstimationQuantization (EQ) algorithm originally designed for still image compression to the ... 
ForWaRD: FourierWavelet Regularized Deconvolution for IllConditioned Systems
(20040201)We propose an efficient, hybrid <i>FourierWavelet Regularized Deconvolution</i> (ForWaRD) algorithm that performs noise regularization via scalar shrinkage in both the Fourier and wavelet domains. The Fourier shrinkage ... 
Hidden Markov Tree Models for Complex Wavelet Transforms
(20020501)Multiresolution models such as the hidden Markov tree (HMT) aim to capture the statistical structure of signals and images by leveraging two key wavelet transform properties: wavelet coefficients representing smooth/singular ... 
Image Data Compression
(19950501)Waveletbased multiresolution representation has become a cuttingedge technology in the area of image data compression. Though the discrete wavelet transform is closely related to the perfectreconstruction octaveband ... 
Image Restoration Using the EM Algorithm and WaveletBased Complexity Regularization
(20020520)This paper introduces an <i>expectationmaximization</i> (EM) algorithm for image restoration (deconvolution) based on a penalized likelihood formulated in the wavelet domain. Regularization is achieved by promoting a ... 
Image SuperResolution for Improved Automatic Target Recognition
(20040401)Infrared imagers used to acquire data for automatic target recognition are inherently limited by the physical properties of their components. Fortunately, image superresolution techniques can be applied to overcome the ... 
JPEG Compression History Estimation for Color Images
(20050701)We routinely encounter digital color images that were previously JPEGcompressed. En route to the image's current representation, the previous JPEG compression's various settings—termed its JPEG compression history ... 
Modelbased Inverse Halftoning with WaveletVaguelette Deconvolution
(20000901)In this paper, we demonstrate based on the linear model of Kite that inverse halftoning is equivalent to the wellstudied problem of deconvolution in the presence of colored noise. We propose the use of the simple and ... 
Multiscale Analysis for Intensity and Density Estimation
(20020420)The nonparametric multiscale polynomial and platelet algorithms presented in this thesis are powerful new tools for signal and image denoising and reconstruction. Unlike traditional waveletbased multiscale methods, these ... 
Multiscale Approximation of Piecewise Smooth TwoDimensional Function using Normal Triangulated Meshes
(20050701)Multiresolution triangulation meshes are widely used in computer graphics for representing threedimensional(3d) shapes. We propose to use these tools to represent 2d piecewise smooth functions such as grayscale ... 
Multiscale Image Segmentation Using Joint Texture and Shape Analysis
(20000701)We develop a general framework to simultaneously exploit texture and shape characterization in multiscale image segmentation. By posing multiscale segmentation as a model selection problem, we invoke the powerful framework ... 
Multiscale Likelihood Analysis and Image Reconstruction
(20030820)The nonparametric multiscale polynomial and platelet methods presented here are powerful new tools for signal and image denoising and reconstruction. Unlike traditional waveletbased multiscale methods, these methods are ... 
Multiscale modeling and estimation of Poisson processes with application to photonlimited imaging
(19990420)Many important problems in engineering and science are wellmodeled by Poisson processes. In many applications it is of great interest to accurately estimate the intensities underlying observed Poisson data. In particular, ... 
On Nearly Orthogonal Lattice Bases
(20050701)We study "nearly orthogonal" lattice bases, or bases where the angle between any basis vector and the linear subspace spanned by the other basis vectors is greater than 60°. We show that a nearly orthogonal lattice ... 
On The Problem of Simultaneous Encoding of Magnitude and Location Information
(20021120)Modern image coders balance bitrate used for encoding the location of signicant transform coefficients, and bitrate used for coding their values. The importance of balancing location and value information in practical ... 
Platelets for Multiscale Analysis in Medical Imaging
(20020420)This paper describes the development and use of multiscale, plateletbased image reconstruction algorithms in medical imaging. Such algorithms are effective because platelets approximate images in certain (piecewise) ... 
Platelets for Multiscale Analysis in PhotonLimited Imaging
(20020920)This paper proposes a new multiscale image decomposition based on platelets. Platelets are localized functions at various scales, locations, and orientations that produce piecewise linear image approximations. For smoothness ... 
Platelets: A Multiscale Approach for Recovering Edges and Surfaces in PhotonLimited Medical Imaging
(20011020)This paper proposes a new multiscale image decomposition based on platelets. Platelets are localized functions at various scales, locations, and orientations that produce piecewise linear image approximations. Platelets ... 
TEMPLAR: A WaveletBased Framework for Pattern Learning and Analysis
(20010420)Despite the success of wavelet decompositions in other areas of statistical signal and image processing, current waveletbased image models are inadequate for modeling patterns in images, due to the presence of unknown ... 
Template Learning from Atomic Representations: A WaveletBased Approach to Pattern Analysis
(20010420)Despite the success of wavelet decompositions in other areas of statistical signal and image processing, current waveletbased image models are inadequate for modeling patterns in images, due to the presence of unknown ... 
Waveletbased deconvolution for illConditioned systems
(20000520)This thesis proposes a new approach to waveletbased image deconvolution that comprises Fourierdomain system inversion followed by waveletdomain noise suppression. In contrast to other waveletbased deconvolution approaches, ... 
Waveletbased Deconvolution for Illconditioned Systems
(19990301)In this paper, we propose a new approach to waveletbased deconvolution. Roughly speaking, the algorithm comprises Fourierdomain system inversion followed by waveletdomain noise suppression. Our approach subsumes a number ... 
Waveletbased deconvolution for illconditioned systems
(20000201)We propose a hybrid approach to waveletbased deconvolution that comprises Fourierdomain system inversion followed by waveletdomain noise suppression. In contrast to other waveletbased deconvolution approaches, the ... 
WaveletBased Deconvolution Using Optimally Regularized Inversion for IllConditioned Systems
(19990720)We propose a hybrid approach to waveletbased deconvolution that comprises Fourierdomain system inversion followed by waveletdomain noise suppression. In contrast to conventional waveletbased deconvolution approaches, ... 
Waveletbased Rician noise removal for magnetic resonance imaging
(19991020)It is wellknown that the noise in magnetic resonance magnitude images obeys a Rician distribution. Unlike additive Gaussian noise, Rician noise is signaldependent and consequently spearating signal from noise is a ... 
WaveletDomain Regularized Deconvolution for IllConditioned Systems
(19991001)We propose a hybrid approach to waveletbased image deconvolution that comprises Fourierdomain system inversion followed by waveletdomain noise suppression. In contrast to conventional waveletbased deconvolution approaches, ... 
WInHD: Waveletbased Inverse Halftoning via Deconvolution
(20021020)We propose the <i>Waveletbased Inverse Halftoning via Deconvolution</i> (WInHD) algorithm to perform inverse halftoning of errordiffused halftones. WInHD is motivated by our realization that inverse halftoning can be ...