Abstract
A new class of interpolation kernels that are locally compact in signal space and "almost band-limited" in Fourier space is presented. The kernels are easy to calculate and lend themselves to problems in which the kernels must be analytically manipulated with other operations or operators such as convolutions and projection integrals. The interpolation kernels are comprised of a linear sum of a Gaussian function and its second derivative (and, when extended to higher order, its higher even derivatives). A numerical Gaussian quadrature method is derived that can be used with integrals involving the kernels that cannot be analytically evaluated. Potential extensions to higher order implementations of the kernels are discussed and examined. The emphasis of the manuscript is on the simplicity of the interpolation kernel and some of its mathematical properties.
Keywords
Interpolation (computer graphics)MathematicsQuadrature (astronomy)Kernel (algebra)Trigonometric interpolationGaussian quadratureGaussianLinear interpolationApplied mathematicsAlgorithmMathematical analysisNearest-neighbor interpolationComputer scienceDiscrete mathematicsIntegral equationArtificial intelligenceNyström method
Affiliated Institutions
Related Publications
Support vector machines for histogram-based image classification
Traditional classification approaches generalize poorly on image classification tasks, because of the high dimensionality of the feature space. This paper shows that support vec...
Computation of electron repulsion integrals using the rys quadrature method
Abstract Following an earlier proposal to evaluate electron repulsion integrals over Gaussian basis functions by a numerical quadrature based on a set of orthogonal polynomials ...
Diffusion Kernels on Graphs and Other Discrete Input Spaces
The application of kernel-based learning algorithms has, so far, largely been confined to realvalued data and a few special data types, such as strings. In this paper we propose...
Invariant Scattering Convolution Networks
A wavelet scattering network computes a translation invariant image representation which is stable to deformations and preserves high-frequency information for classification. I...
Sampling signals with finite rate of innovation
Consider classes of signals that have a finite number of degrees of freedom per unit of time and call this number the rate of innovation. Examples of signals with a finite rate ...
Publication Info
- Year
- 1996
- Type
- article
- Volume
- 15
- Issue
- 3
- Pages
- 369-376
- Citations
- 45
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
45
OpenAlex
Cite This
C. Robert Appledorn
(1996).
A new approach to the interpolation of sampled data.
IEEE Transactions on Medical Imaging
, 15
(3)
, 369-376.
https://doi.org/10.1109/42.500145
Identifiers
- DOI
- 10.1109/42.500145