Search results for "Total variation"

showing 10 items of 26 documents

Diffusion map for clustering fMRI spatial maps extracted by Indipendent Component Analysis

2013

Functional magnetic resonance imaging (fMRI) produces data about activity inside the brain, from which spatial maps can be extracted by independent component analysis (ICA). In datasets, there are n spatial maps that contain p voxels. The number of voxels is very high compared to the number of analyzed spatial maps. Clustering of the spatial maps is usually based on correlation matrices. This usually works well, although such a similarity matrix inherently can explain only a certain amount of the total variance contained in the high-dimensional data where n is relatively small but p is large. For high-dimensional space, it is reasonable to perform dimensionality reduction before clustering.…

FOS: Computer and information sciencesDiffusion (acoustics)Computer sciencediffusion mapMachine Learning (stat.ML)02 engineering and technologycomputer.software_genreMachine Learning (cs.LG)Computational Engineering Finance and Science (cs.CE)Correlation03 medical and health sciencesTotal variation0302 clinical medicineStatistics - Machine LearningVoxel0202 electrical engineering electronic engineering information engineeringComputer Science - Computational Engineering Finance and ScienceCluster analysisdimensionality reductionta113spatial mapsbusiness.industryDimensionality reductionfunctional magnetic resonance imaging (fMRI)Pattern recognitionIndependent component analysisSpectral clusteringComputer Science - Learningindependent component analysista6131020201 artificial intelligence & image processingArtificial intelligenceDYNAMICAL-SYSTEMSbusinesscomputer030217 neurology & neurosurgeryclustering
researchProduct

Denoising of MR spectroscopy signals using total variation and iterative Gauss-Seidel gradient updates

2015

We present a fast variational approach for denoising signals from magnetic resonance spectroscopy (MRS). Differently from the TV approaches applied to denoising of images, this is the first time to our knowledge that it has been used for the processing of free induction decay signals from single-voxel spectroscopy (SVS) acquisitions. Another novelty in this study is the direct use of the Euler Lagrange formulation coupled with Gauss Seidel gradient updates to improve the speed of iteration and reduce ringing. Results from brain MRS signals show improvement in signal to noise ratio as well as reduction in estimation error in the quantification of metabolites.

Free induction decayReduction (complexity)Mathematical optimizationSignal-to-noise ratioNoise reductionGauss–Seidel methodRingingTotal variation denoisingSpectroscopyAlgorithmMathematics2015 IEEE 12th International Symposium on Biomedical Imaging (ISBI)
researchProduct

A ‘TVD-like’ Scheme for Conservation Laws with Source Terms

2008

The theoretical foundations of high-resolution TVD schemes for homogeneous scalar conservation laws and linear systems of conservation laws have been firmly established through the work of Harten [5], Sweby [11], and Roe [9]. These TVD schemes seek to prevent an increase in the total variation of the numerical solution, and are successfully implemented in the form of flux-limiters or slope limiters for scalar conservation laws and systems. However, their application to conservation laws with source terms is still not fully developed. In this work we analyze the properties of a second order, flux-limited version of the Lax-Wendroff scheme preserving steady states [3]. Our technique is based …

Fully developedFlux limitingConservation lawHomogeneousTotal variation diminishingLinear systemScalar (mathematics)Applied mathematicsHigh-resolution schemeMathematics
researchProduct

Fast Image Restoration Algorithms Based on PDE Models Using Modified Hopfield Neural Network

2010

Two image restoration algorithms based on modified Hop field neural network and variational partial differential equations (PDE) were proposed in our previous work [1, 2]. But the convergence rate of the proposed algorithms was slow. In this paper, we develop a fast update rule based on modified Hop field neural network (MHNN) of continuous state change and two fast image restoration algorithms. Experimental results show that, when compared with the previous algorithms, our proposed algorithms have better performance both in convergence rate and in image restoration quality.

Harmonic analysisPartial differential equationArtificial neural networkRate of convergenceComputer scienceSignal processing algorithmsTotal variation modelRule-based systemAlgorithmImage restoration2010 International Conference on Artificial Intelligence and Computational Intelligence
researchProduct

A note on the Bregmanized Total Variation and dual forms

2009

This paper considers two approaches to perform image restoration while preserving the contrast. The first one is the Total Variation-based Bregman iterations while the second consists in the minimization of an energy that involves robust edge preserving regularization. We show that these two approaches can be derived form a common framework. This allows us to deduce new properties and to extend and generalize these two previous approaches.

Mathematical optimizationNoise measurementIterative methodCommon frameworkMinificationTotal variation denoisingAlgorithmRegularization (mathematics)Image restorationMathematics2009 16th IEEE International Conference on Image Processing (ICIP)
researchProduct

Subsignal-based denoising from piecewise linear or constant signal

2011

15 pages; International audience; n the present work, a novel signal denoising technique for piecewise constant or linear signals is presented termed as "signal split." The proposed method separates the sharp edges or transitions from the noise elements by splitting the signal into different parts. Unlike many noise removal techniques, the method works only in the nonorthogonal domain. The new method utilizes Stein unbiased risk estimate (SURE) to split the signal, Lipschitz exponents to identify noise elements, and a polynomial fitting approach for the sub signal reconstruction. At the final stage, merging of all parts yield in the fully denoised signal at a very low computational cost. St…

Mathematical optimization[ INFO.INFO-TS ] Computer Science [cs]/Signal and Image ProcessingComputer scienceStochastic resonanceNoise reduction[ SPI.SIGNAL ] Engineering Sciences [physics]/Signal and Image processing02 engineering and technology01 natural sciencesMultiplicative noisePiecewise linear function010104 statistics & probabilitySpeckle patternsymbols.namesakeSignal-to-noise ratioWavelet[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing0202 electrical engineering electronic engineering information engineering0101 mathematicsSignal transfer functionShrinkageSignal reconstructionNoise (signal processing)General EngineeringNonlinear opticsWavelet transform020206 networking & telecommunicationsTotal variation denoisingAtomic and Molecular Physics and OpticsAdditive white Gaussian noiseGaussian noisePiecewisesymbolsStep detectionAlgorithm[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing
researchProduct

Local and nonlocal weighted pLaplacian evolution equations with Neumann boundary conditions

2011

In this paper we study existence and uniqueness of solutions to the local diffusion equation with Neumann boundary conditions and a bounded nonhomogeneous diffusion coefficient g ≥ 0, {ut = div (g|∇u|p-2∇u) in ]0; T[×Ωg|∇u|p-2u·n = 0 on ]0; T[×∂Ω; for 1 ≤ p < ∞. We show that a nonlocal counterpart of this diffusion problem is ut(t; x)= ∫ω J(x-y)g(x+y/2)|u(t; y)-u(t; x)| p-2 (u(t; y)-u(t; x)) dy in ]0; T[× Ω,where the diffusion coefficient has been reinterpreted by means of the values of g at the point x+y/2 in the integral operator. The fact that g ≥ 0 is allowed to vanish in a set of positive measure involves subtle difficulties, specially in the case p = 1.

Neumann boundary conditionsDiffusion equationGeneral MathematicsOperator (physics)Nonlocal diffusionMathematical analysisMeasure (mathematics)P-laplacianBounded functionNeumann boundary conditionp-LaplacianUniquenessDiffusion (business)Total variation flowMathematicsMathematical physics
researchProduct

A nonlocal p-Laplacian evolution equation with Neumann boundary conditions

2008

In this paper we study the nonlocal p-Laplacian type diffusion equation,ut (t, x) = under(∫, Ω) J (x - y) | u (t, y) - u (t, x) |p - 2 (u (t, y) - u (t, x)) d y . If p &gt; 1, this is the nonlocal analogous problem to the well-known local p-Laplacian evolution equation ut = div (| ∇ u |p - 2 ∇ u) with homogeneous Neumann boundary conditions. We prove existence and uniqueness of a strong solution, and if the kernel J is rescaled in an appropriate way, we show that the solutions to the corresponding nonlocal problems converge strongly in L∞ (0, T ; Lp (Ω)) to the solution of the p-Laplacian with homogeneous Neumann boundary conditions. The extreme case p = 1, that is, the nonlocal analogous t…

Neumann boundary conditionsMathematics(all)Diffusion equationApplied MathematicsGeneral MathematicsNonlocal diffusionMathematical analysisp-LaplacianFlow (mathematics)Neumann boundary conditionp-LaplacianInitial value problemUniquenessBoundary value problemCalculus of variationsTotal variation flowMathematicsJournal de Mathématiques Pures et Appliquées
researchProduct

Total-variation-based methods for gravitational wave denoising

2014

We describe new methods for denoising and detection of gravitational waves embedded in additive Gaussian noise. The methods are based on Total Variation denoising algorithms. These algorithms, which do not need any a priori information about the signals, have been originally developed and fully tested in the context of image processing. To illustrate the capabilities of our methods we apply them to two different types of numerically-simulated gravitational wave signals, namely bursts produced from the core collapse of rotating stars and waveforms from binary black hole mergers. We explore the parameter space of the methods to find the set of values best suited for denoising gravitational wa…

PhysicsNuclear and High Energy PhysicsGravitational waveNoise (signal processing)Noise reductionFOS: Physical sciencesImage processingGeneral Relativity and Quantum Cosmology (gr-qc)Total variation denoisingGeneral Relativity and Quantum Cosmologysymbols.namesakeClassical mechanicsBinary black holeGaussian noisesymbolsWaveformAstrophysics - Instrumentation and Methods for AstrophysicsInstrumentation and Methods for Astrophysics (astro-ph.IM)AlgorithmPhysical Review D
researchProduct

Large deviations results for subexponential tails, with applications to insurance risk

1996

AbstractConsider a random walk or Lévy process {St} and let τ(u) = inf {t⩾0 : St > u}, P(u)(·) = P(· | τ(u) < ∞). Assuming that the upwards jumps are heavy-tailed, say subexponential (e.g. Pareto, Weibull or lognormal), the asymptotic form of the P(u)-distribution of the process {St} up to time τ(u) is described as u → ∞. Essentially, the results confirm the folklore that level crossing occurs as result of one big jump. Particular sharp conclusions are obtained for downwards skip-free processes like the classical compound Poisson insurance risk process where the formulation is in terms of total variation convergence. The ideas of the proof involve excursions and path decompositions for Mark…

Statistics and ProbabilityExponential distributionRegular variationRuin probabilityExcursionRandom walkDownwards skip-free processLévy processConditioned limit theoremTotal variation convergenceCombinatoricsInsurance riskPath decompositionIntegrated tailProbability theoryModelling and SimulationExtreme value theoryMaximum domain of attractionMathematicsStochastic processApplied MathematicsExtreme value theoryRandom walkSubexponential distributionModeling and SimulationLog-normal distributionLarge deviations theory60K1060F10Stochastic Processes and their Applications
researchProduct