Numerical Algorithms

, Volume 10, Issue 2, pp 401–419 | Cite as

Convolution kernels based on thin-plate splines

  • J. Levesley


Quasi-interpolation using radial basis functions has become a popular method for constructing approximations to continuous functions in many space dimensions. In this paper we discuss a procedure for generating kernels for quasi-interpolation, using functions which have series expansions involving terms likerα logr. It is shown that such functions are suitable if and only if α is a positive even integer and the spatial dimension is also even.


Radial basis functions thin-plate splines convolution kernels 

AMS(MOS) subject classification

41A30 41A63 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    M. Abramowitz and I.A. Stegun,Handbook of Mathematical Functions, National Bureau of Standards (Dover, 1964).Google Scholar
  2. [2]
    M.D. Buhmann and M.J.D. Powell, Radial basis function interpolation on an infinite regular grid, DAMTP 1988/NA13 (1988).Google Scholar
  3. [3]
    I.R.H. Jackson, Convergence properties of radial basis functions, Constr. Approx. 4 (1988) 243–264.CrossRefGoogle Scholar
  4. [4]
    F. John,Plane Waves and Spherical Means Applied to Partial Differential Equations (Interscience, New York, 1955).Google Scholar
  5. [5]
    D.S. Jones,Generalised Functions (McGraw-Hill, London, 1966).Google Scholar
  6. [6]
    J. Levesley, Y. Xu, W.A. Light and E.W. Cheney, Convolution kernels for radial basis function approximation, SIAM J. Math. Anal., to be published.Google Scholar
  7. [7]
    W.A. Light, private communication (May 1992).Google Scholar
  8. [8]
    M.J.D. Powell, The theory of radial basis function approximation in 1900, in:Advances in Numerical Analysis, Vol. 2, ed. W.A. Light (Clarendon Press, Oxford, 1991).Google Scholar
  9. [9]
    W. Rudin,Real and Complex Analysis (McGraw-Hill, 1966).Google Scholar

Copyright information

© J.C. Baltzer AG, Science Publishers 1995

Authors and Affiliations

  • J. Levesley
    • 1
  1. 1.Department of Mathematics and Computer ScienceUniversity of LeicesterLeicesterUK

Personalised recommendations