Image Decomposition by Radial Basis Functions

  • Jens D. Andersen
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2749)


In order to reconcile linear and morphological scale space an image decomposition based on radial basis functions (RBF’s) may be used. Radial-basis functions can be used to synthesize approximations of multidimensional functions, i.e. solving the problem of hypersurface reconstruction thus approximating the image intensities. They can be used in several ways in RBF networks which are linear neural networks. They provide a link between linear and morphological scale spaces. A morphological scale-space is a scale dependent decomposition of images from coarse to fine scale based on morphological operations on images. This is in contrast to the linear scale space, which is based on gaussian smoothing of image features. Both types of scale space are being advocated for image segmentation.


Scale space radial basis functions neural networks 


  1. 1.
    L. Alvarez, F. Guichard, P.-L. Lions and J.-M. Morel: Axioms and fundamental equations of image processing. Arch. Rational Mech. Anal. vol. 123, pp. 199–257, 1993.zbMATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    I. Fujita, K. Tanaka, M. Ito and K. Cheng, Columns for visual features of objects in monkey inferotemporal cortex, Nature, vol. 360, pp. 343–346, november 1992.CrossRefGoogle Scholar
  3. 3.
    H.J.A.M. Heijmans and R.v.d. Boomgaard: Algebraic Framework for linear and morphological scale-spaces. Journal of Visual Communication and Image Representation, vol. 13(1/2), pp. 269–301, March 2002.CrossRefGoogle Scholar
  4. 4.
    J. Froment: A compact and multiscale image model based on level sets. In: Scale-Space’ 99 (M. Nielsen et al, eds.), Berlin: Springer-Verlag: LNCS 1682, pp. 152–163, 1999.Google Scholar
  5. 5.
    T. Lindeberg: Scale-space theory in computer vision. Doordrect, Kluwer Academic Publishers 1994.Google Scholar
  6. 6.
    F. Meyer and P. Maragos: Morphological scale-space representation with levelings. In: Scale-Space’ 99 (M. Nielsen et al, eds.), Berlin: Springer-Verlag: LNCS 1682, pp. 187–198, 1999.Google Scholar
  7. 7.
    Mark J.L. Orr: Introduction to Radial Basis Function Networks. Technical Report, Centre for Cognitive Science, University of Edinburgh 1996. Scholar
  8. 8.
    O.F. Olsen: Generic image structure. Ph.D.-thesis. Department of Computer Science, Faculty of Science, University of Copenhagen 2000.Google Scholar
  9. 9.
    O.F. Olsen and M. Nielsen: Generic events for the gradient squared with application to multi-scale segmentation. In: B. ter Haar Romeny et al: Scale-Space Theory in Computer Vision 1997, pp. 101–112, Berlin: Springer-Verlag: LNCS 1252, 1997.Google Scholar
  10. 10.
    T. Poggio and F. Girosi, Networks for Approximation and Learning, Proc. IEEE, Vol. 78, No. 9, pp. 1481–1497, September 1990.CrossRefGoogle Scholar
  11. 11.
    J. Sporring et al.: Gaussian scale-space theory. Doordrect, Kluwer Academic Publishers 1997.zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Jens D. Andersen
    • 1
  1. 1.Department of Computer ScienceUniversitetsparken 1Copenhagen Ø

Personalised recommendations