Journal of Mathematical Imaging and Vision

, Volume 7, Issue 4, pp 291–307 | Cite as

Regularization, Scale-Space, and Edge Detection Filters

  • Mads Nielsen
  • Luc Florack
  • Rachid Deriche


Computational vision often needs to deal with derivatives ofdigital images. Such derivatives are not intrinsic properties ofdigital data; a paradigm is required to make them well-defined.Normally, a linear filtering is applied. This can be formulated interms of scale-space, functional minimization, or edge detectionfilters. The main emphasis of this paper is to connect these theoriesin order to gain insight in their similarities and differences. We donot want, in this paper, to take part in any discussion of how edgedetection must be performed, but will only link some of the current theories. We take regularization (or functional minimization) as astarting point, and show that it boils down to Gaussian scale-space ifwe require scale invariance and a semi-group constraint to besatisfied. This regularization implies the minimization of afunctional containing terms up to infinite order of differentiation.If the functional is truncated at second order, the Canny-Deriche filter arises. It is also shown that higher dimensional regularizationboils down to a rotated version of the one dimensional case, whenCartesian invariance is imposed and the image is vanishing at theborders. This means that the results from 1D regularization can beeasily generalized to higher dimensions. Finally we show how anefficient implementation of regularization of order n can be made byrecursive filtering using 2n multiplications and additions peroutput element without introducing any approximation.

regularization linear filtering scale invariance scale-space recursive filtering edge detection filters 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    L. Schwartz, “Theorie des distributions,” Actualités scientifigues et industrielles, Publications de l’Institut de Mathématique de l’Université de Strasbourg, Vols. I, II, 1950–1951.Google Scholar
  2. 2.
    A.P. Witkin, “Scale-space filtering,” Proc. of IJCAI, Karlsruhe, Germany, 1983.Google Scholar
  3. 3.
    J.J. Koenderink, “The structure of images,” Biol. Cybern., Vol. 50, pp. 363–370, 1984.Google Scholar
  4. 4.
    L. Florack, “The syntactical structure of scalar images,” Ph.D. thesis, University of Utrecht, 1993.Google Scholar
  5. 5.
    A.N. Tikhonov and V.Y. Arseninn, Solution of Ill-posed Problems, V.H. Winston & Sons, John Wiley, 1977.Google Scholar
  6. 6.
    T. Poggio, H. Voorhees, and A. Yuille, “A regularized solution to edge detection,” A.I. Memo 883, M.I.T, May 1985.Google Scholar
  7. 7.
    J.F. Canny, “A computational approach to edge detection,” IEEE PAMI, Vol. 8, No.6, Nov. 1986.Google Scholar
  8. 8.
    R. Deriche, “Using Canny's criteria to derive a recursively implemented optimal edge detector,” Int. Journal of Computer Vision, Vol. 1, No.2, pp. 167–187, 1987.Google Scholar
  9. 9.
    M. Nielsen, L. Florack, and R. Deriche, “Regularization, scalespace, and edge detection filters,” Proc. of ECCV 96, Cambridge, UK, April 14–18, 1996, pp. 70–81.Google Scholar
  10. 10.
    J. Fourier, The Analytic Theory of Heat, New York: Dover Publications, 1995, Original Théori Analytique de la Chaleur, Paris, 1822.Google Scholar
  11. 11.
    T. Lindeberg, Scale-Space Theory in Computer Vision, Kluwer Academic Publishers: Netherlands, 1994.Google Scholar
  12. 12.
    T. Lindeberg, “On the axiomatic foundations of linear scalespace: Combining semi-group structure with causality vs. scale invariance,” Technical report, CVAP-159, Royal Institute of Technology, Stockholm, Sweden, 1994.Google Scholar
  13. 13.
    E.J. Pauwels, P. Fiddelaers, T. Moons, and L.J.V. Gool, “An extended class of scale-invariant and recursive scale space filters,” Katholieke Universiteit Leuven, Tech. Rep. KUL/ESAT/MI2/9316, 1994.Google Scholar
  14. 14.
    D. Hilbert, “Ueber die vollen invariantsystemen,” Math. Annalen, Vol. 42, pp. 313–373, 1893.Google Scholar
  15. 15.
    W.E.L. Grimson, From Images to Surfaces, MIT Press: Cambridge, Massachusetts, 1981.Google Scholar
  16. 16.
    R. Deriche, “Fast algorithms for low-level vision,” IEEE PAMI, Vol. 12, No.1, pp. 78–88, Jan. 1990.Google Scholar
  17. 17.
    R. Deriche, “Recursively implementing the Gaussian and its derivatives,” Proc. 2nd Int. Conference on Image Processing, Singapore, Sept. 7–11, 1992, pp. 263–267.Google Scholar
  18. 18.
    S. Castan, J. Zhao, and J. Shen, “Optimal filter for edge detection methods and results,” ECCV 90, Antibes, France, May 1990.Google Scholar
  19. 19.
    M. Unser, A. Aldroubi, and M. Eden, “Recursive regularization filters: Design, properties and applications,” IEEE PAMI, Vol. 13, No.3, March 1991.Google Scholar
  20. 20.
    M. Nielsen, L. Florack, and R. Deriche, “Regularization and scale-space,” Research Report 2352, INRIA, Sophia-Antipolis, France, 1994, URL: Scholar
  21. 21.
    P. Perona and J. Malik, “Scale space and edge detection using anisotropic diffusion,” IEEE PAMI, Vol. 12, pp. 429–439, 1990.Google Scholar
  22. 22.
    N. Nordström, “Biased anisotropic diffusion—A unified regularization and diffusion approach to edge detection,” Image and Visual Computing, No. 8, pp. 318–327, 1990.Google Scholar

Copyright information

© Kluwer Academic Publishers 1997

Authors and Affiliations

  • Mads Nielsen
    • 1
  • Luc Florack
    • 1
  • Rachid Deriche
    • 2
  1. 1.DIKU, Universitetsparken 1CopenhagenDenmark
  2. 2.INRIASophia Antipolis CedexFrance

Personalised recommendations