%0 Journal Article %J IEEE Transactions on Image Processing %D 2012 %T Gradient-based Image Recovery Methods from Incomplete Fourier Measurements %A Patel, Vishal M. %A Maleh,R. %A Gilbert,A. C %A Chellapa, Rama %K Compressed sensing %K Fourier transforms %K Image coding %K Image edge detection %K Image reconstruction %K L1–minimization %K minimization %K Noise measurement %K OPTIMIZATION %K Poisson solver %K Sparse recovery %K Total variation %K TV %X A major problem in imaging applications such as Magnetic Resonance Imaging (MRI) and Synthetic Aperture Radar (SAR) is the task of trying to reconstruct an image with the smallest possible set of Fourier samples, every single one of which has a potential time and/or power cost. The theory of Compressive Sensing (CS) points to ways of exploiting inherent sparsity in such images in order to achieve accurate recovery using sub- Nyquist sampling schemes. Traditional CS approaches to this problem consist of solving total-variation minimization programs with Fourier measurement constraints or other variations thereof. This paper takes a different approach: Since the horizontal and vertical differences of a medical image are each more sparse or compressible than the corresponding total-variational image, CS methods will be more successful in recovering these differences individually. We develop an algorithm called GradientRec that uses a CS algorithm to recover the horizontal and vertical gradients and then estimates the original image from these gradients. We present two methods of solving the latter inverse problem: one based on least squares optimization and the other based on a generalized Poisson solver. After a thorough derivation of our complete algorithm, we present the results of various experiments that compare the effectiveness of the proposed method against other leading methods. %B IEEE Transactions on Image Processing %V PP %P 1 - 1 %8 2012/// %@ 1057-7149 %G eng %N 99 %R 10.1109/TIP.2011.2159803 %0 Conference Paper %B 2011 IEEE International Conference on Computer Vision (ICCV) %D 2011 %T Blurring-invariant Riemannian metrics for comparing signals and images %A Zhengwu Zhang %A Klassen, E. %A Srivastava, A. %A Turaga,P. %A Chellapa, Rama %K blurring-invariant Riemannian metrics %K Estimation %K Fourier transforms %K Gaussian blur function %K Gaussian processes %K image representation %K log-Fourier representation %K measurement %K Orbits %K Polynomials %K signal representation %K Space vehicles %K vectors %X We propose a novel Riemannian framework for comparing signals and images in a manner that is invariant to their levels of blur. This framework uses a log-Fourier representation of signals/images in which the set of all possible Gaussian blurs of a signal, i.e. its orbits under semigroup action of Gaussian blur functions, is a straight line. Using a set of Riemannian metrics under which the group actions are by isometries, the orbits are compared via distances between orbits. We demonstrate this framework using a number of experimental results involving 1D signals and 2D images. %B 2011 IEEE International Conference on Computer Vision (ICCV) %I IEEE %P 1770 - 1775 %8 2011/11/06/13 %@ 978-1-4577-1101-5 %G eng %R 10.1109/ICCV.2011.6126442 %0 Conference Paper %B IEEE International Conference on Acoustics, Speech and Signal Processing, 2008. ICASSP 2008 %D 2008 %T Measuring 1st order stretchwith a single filter %A Bitsakos,K. %A Domke, J. %A Fermüller, Cornelia %A Aloimonos, J. %K Cepstral analysis %K Educational institutions %K filter %K filtering theory %K Fourier transforms %K Frequency domain analysis %K Frequency estimation %K Gabor filters %K Image analysis %K IMAGE PROCESSING %K linear stretch measurement %K local signal transformation measurement %K Nonlinear filters %K Phase estimation %K Signal analysis %K Speech processing %X We analytically develop a filter that is able to measure the linear stretch of the transformation around a point, and present results of applying it to real signals. We show that this method is a real-time alternative solution for measuring local signal transformations. Experimentally, this method can accurately measure stretch, however, it is sensitive to shift. %B IEEE International Conference on Acoustics, Speech and Signal Processing, 2008. ICASSP 2008 %I IEEE %P 909 - 912 %8 2008/04/31/March %@ 978-1-4244-1483-3 %G eng %R 10.1109/ICASSP.2008.4517758 %0 Conference Paper %B IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2005 %D 2005 %T Plane-wave decomposition analysis for spherical microphone arrays %A Duraiswami, Ramani %A Zhiyun Li %A Zotkin,Dmitry N %A Grassi,E. %A Gumerov, Nail A. %K Acoustic propagation %K Acoustic scattering %K acoustic signal processing %K acoustic waves %K array signal processing %K band-limit criteria %K beamforming %K Educational institutions %K Fourier transforms %K Frequency %K Laboratories %K microphone arrays %K Nails %K Nyquist criterion %K Nyquist-like criterion %K Partial differential equations %K plane-wave decomposition analysis %K sound field analysis %K spherical microphone arrays %K spherical wave-functions %X Spherical microphone arrays have attracted attention for analyzing the sound field in a region and beamforming. The analysis of the recorded sound has been performed in terms of spherical wave-functions, and recently the use of plane-wave expansions has been suggested. We show that the plane-wave basis is intimately related to the spherical wave-functions. Reproduction in terms of both representations satisfies certain band-limit criteria. We provide an error bound that shows that to reproduce the spatial characteristics of a sound of a certain frequency we need to be able to accurately represent sounds of up to a particular order, which establishes a Nyquist-like criterion. The order of the sound field in turn is related to the number of microphones in the array necessary to achieve accurate quadrature on the sphere. These results are illustrated with simulations. %B IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2005 %I IEEE %P 150 - 153 %8 2005/10// %@ 0-7803-9154-3 %G eng %R 10.1109/ASPAA.2005.1540191 %0 Conference Paper %B Automation Congress, 2002 Proceedings of the 5th Biannual World %D 2002 %T Hidden Markov models for silhouette classification %A Abd-Almageed, Wael %A Smith,C. %K Computer vision %K Feature extraction %K Fourier transforms %K hidden Markov models %K HMM %K image classification %K Neural networks %K object classification %K Object recognition %K parameter estimation %K pattern recognition %K Probability distribution %K Shape measurement %K silhouette classification %K Wavelet transforms %X In this paper, a new technique for object classification from silhouettes is presented. Hidden Markov models are used as a classification mechanism. Through a set of experiments, we show the validity of our approach and show its invariance under severe rotation conditions. Also, a comparison with other techniques that use hidden Markov models for object classification from silhouettes is presented. %B Automation Congress, 2002 Proceedings of the 5th Biannual World %I IEEE %V 13 %P 395 - 402 %8 2002/// %@ 1-889335-18-5 %G eng %R 10.1109/WAC.2002.1049575