Signal, Image and Video Processing

, Volume 1, Issue 3, pp 273–285 | Cite as

Statistical behavior of edge detectors

Original paper

Abstract

In this paper we present a method for estimating the statistical properties of two well-known edge detectors: the non maxima suppression and the zero crossing of the Laplacian algorithms. Assuming the data are corrupted by an additive Gaussian noise we derive the probability density function (pdf) of the detected edge. Thanks to this approach the computed pdf explicitly depends on the parameters of the edge detector. Experimental results on real images and comparisons with Monte Carlo simulations are presented in order to characterize the performance of this method.

Keywords

Edge detector Statistical model 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Roberts, L.G.: Machine perception of three-dimensional solids. In: Optical and Electro-Optical Information Processing. MIT Press, Cambridge (1965)Google Scholar
  2. 2.
    Sobel, I.: Camera models and machine perception. Standford AI MEMO 121, Dept. of Computer Science, Standford University (1970)Google Scholar
  3. 3.
    Gonzalez R. and Woods R. (1992). Digital Image Processing. Addison-Wesley, New York Google Scholar
  4. 4.
    Canny J.F. (1986). A computational approach to edge detection. IEEE T. PAMI 8(6): 679–698 Google Scholar
  5. 5.
    Deriche R. (1987). Using Canny’s criteria to derive a recursively implemented optimal edge detector. Int. J. Comp. Vision 1(2): 167–187 CrossRefGoogle Scholar
  6. 6.
    Shen J. and Castan S. (1992). An optimal linear operator for step edge detection. Comp. Vision Graph. Image Process. 54(2): 112–133 Google Scholar
  7. 7.
    Demigny D. (2002). An optimal linear filtering for edge detection. IEEE Trans. Image Process. 11(7): 728–737 CrossRefGoogle Scholar
  8. 8.
    Rosenfeld A., Hummel R. and Zucker S. (1976). Scenes labelling by relaxation operation. IEEE Trans. Systems Man Cyber. 6(6): 420–453 MATHMathSciNetCrossRefGoogle Scholar
  9. 9.
    Hancock E.R. and Kittler J. (1990). Edge-labeling using dictionary relaxation. IEEE Trans. PAMI 12: 165–181 Google Scholar
  10. 10.
    Christmas W., Kittler J. and Petrou M. (1995). Structural matching in computer vision using probabilistic relaxation. IEEE Trans. PAMI 17: 749–764 Google Scholar
  11. 11.
    Sitharama Iyengar S. and Deng W. (1995). An efficient edge detection algorithm using relaxation labeling technique. Pattern Recogn. 28(4): 519–536 CrossRefGoogle Scholar
  12. 12.
    Papachistous P., Petrou M. and Kittler J. (2000). Edge postprocessing using probabilistic relaxation. IEEE Trans. Systems Man Cybern. B 30(3): 383–402 CrossRefGoogle Scholar
  13. 13.
    Lyvers E.P. and Mitchell O.R. (1988). Precision edge contrast and orientation estimation. IEEE Trans. PAMI 10(6): 927–937 Google Scholar
  14. 14.
    Marimont D.H., Rubner, Y.: A Probabilistic framework for edge detection and scale selection. In: Proc. of Sixth Int. Conf. Computer Vision, Bombay, pp. 207–214 (1998)Google Scholar
  15. 15.
    Touzi R., Lopes A. and Bousquet P. (1988). A statistical and geometrical edge detector for SAR images. IEEE Trans. Geosci. Remote Sensing 26(6): 764–773 CrossRefGoogle Scholar
  16. 16.
    Tupin F., Maître H., Mangin J.-F., Nicolas J.-M. and Pechersky E. (1998). Detection of linear features in SAR images: application to road network extraction. IEEE Trans. Geosci. Remote Sensing 36(2): 434–453 CrossRefGoogle Scholar
  17. 17.
    Destrempes F. and Mignotte M. (2004). A statistical model for contours in images. IEEE Trans. PAMI 25(5): 626–640 Google Scholar
  18. 18.
    Konishi S., Yuille A.L., Coughlan J.M. and Chun Zhu S. (2003). Statistical edge detection: learning and evaluating edge cues. IEEE Trans. PAMI 25(1): 57–74 Google Scholar
  19. 19.
    Papoulis A. (1991). Probability, Random Variables and Stochastic Processes, 3rd edn. McGraw-Hill, New York Google Scholar
  20. 20.
    Ramdani S., Bouchara F. and Djemal K. (2004). Probability distribution of sub-pixel edge position. Signal Process. 83(3): 445–452 CrossRefGoogle Scholar
  21. 21.
    Genz A. (1992). Numerical computation of multivariate normal probabilities. J. Comput. Graph. Statist. 1(2): 141–149 CrossRefGoogle Scholar
  22. 22.
    Sheil J. and O’Muircheartaigh I. (1977). Algorithm AS 106: the distribution of non-negative quadratic forms in normal variables. Appl. Statist. 26(1): 92–98 CrossRefGoogle Scholar
  23. 23.
    Bouchara F. (2004). Efficient algorithm for the computation of the second order moment of the subpixel-edge position. Appl. Opt. 43(23): 4550–4558 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2007

Authors and Affiliations

  1. 1.UMR CNRS 6168 LSISUniversité du Sud Toulon-VarLa Garde CedexFrance
  2. 2.EA 2991 EDM, Université de Montpellier IMontpellierFrance

Personalised recommendations