Search results for "Upsampling"

showing 5 items of 5 documents

FAST EDGE-FILTERED IMAGE UPSAMPLING.

2011

We present a novel edge preserved interpolation scheme for fast upsampling of natural images. The proposed piecewise hyperbolic operator uses a slope-limiter function that conveniently lends itself to higher-order approximations and is responsible for restricting spatial oscillations arising due to the edges and sharp details in the image. As a consequence the upsampled image not only exhibits enhanced edges, and discontinuities across boundaries, but also preserves smoothly varying features in images. Experimental results show an improvement in the PSNR compared to typical cubic, and spline-based interpolation approaches.

business.industryIterative reconstructionClassification of discontinuitiesEdge detectionArticleUpsamplingSpline (mathematics)PiecewiseComputer visionArtificial intelligenceFlux limiterbusinessImage resolutionMathematicsProceedings. International Conference on Image Processing
researchProduct

Non-periodic Discrete Splines

2015

Discrete Splines with different spans were introduced in Sect. 3.3.1. This chapter focuses on a special case of discrete splines whose spans are powers of 2. These splines are discussed in more detail. The Zak transform provides an integral representation of such splines. Discrete exponential splines are introduced. Generators of the discrete-spline spaces are described whose properties are similar to properties of polynomial-spline spaces generators. Interpolating discrete splines provide efficient tools for upsampling 1D and 2D signals. An algorithm for explicit computation of discrete splines is described.

UpsamplingComputer Science::GraphicsIntegral representationCharacteristic function (probability theory)ComputationZak transformApplied mathematicsSpecial caseInfinite impulse responseFourier seriesMathematics::Numerical AnalysisMathematics
researchProduct

Fast Computation by Subdivision of Multidimensional Splines and Their Applications

2016

We present theory and algorithms for fast explicit computations of uni- and multi-dimensional periodic splines of arbitrary order at triadic rational points and of splines of even order at diadic rational points. The algorithms use the forward and the inverse Fast Fourier transform (FFT). The implementation is as fast as FFT computation. The algorithms are based on binary and ternary subdivision of splines. Interpolating and smoothing splines are used for a sample rate convertor such as resolution upsampling of discrete-time signals and digital images and restoration of decimated images that were contaminated by noise. The performance of the rate conversion based spline is compared with the…

interpolating and smoothing splinesComputer Science::Graphicsrestorationprolate spheroidal wave functionsrate convertorperiodic splinessubdivisionupsamplingMathematicsofComputing_NUMERICALANALYSISComputingMethodologies_COMPUTERGRAPHICS
researchProduct

Non-periodic Discrete-Spline Wavelets

2015

This chapter describes wavelet analysis in the spaces of discrete splines whose spans are powers of 2. This wavelet analysis is similar to wavelet analysis in the polynomial-spline spaces. The transforms are based on relations between exponential discrete splines from different resolution scales. Generators of discrete-spline wavelet spaces are described. The discrete-spline wavelet transforms generate wavelet transforms in signal space. Practically, wavelet transforms of signals are implemented by multirate filtering of signals by two-channel filter banks with the downsampling factor 2 (critically sampled filter banks). The filtering implementation is accelerated by switching to the polyph…

UpsamplingSpline (mathematics)WaveletComputer scienceComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONMathematicsofComputing_NUMERICALANALYSISWavelet transformPolyphase systemData_CODINGANDINFORMATIONTHEORYFilter (signal processing)AlgorithmComputingMethodologies_COMPUTERGRAPHICSExponential function
researchProduct

Splines Computation by Subdivision

2015

In this chapter, fast stable algorithms are presented, which compute splines’ values at dyadic and triadic rational points starting from their samples at integer grid points. The algorithms are implemented by the causal-anticausal recursive filtering of initial data samples, which is followed by iterated application of FIR filters. Extension of the algorithms to the multidimensional case is straightforward. A natural application of the presented subdivision algorithms is for upsampling of signals and images. A few upsampling examples are provided.

UpsamplingBox splineFinite impulse responseComputer scienceIterated functionbusiness.industryComputationExtension (predicate logic)businessAlgorithmInteger gridSubdivision
researchProduct