Advertisement

Visual Form pp 333-344 | Cite as

Entropy Scale-Space

  • Benjamin B. Kimia
  • Allen Tannenbaum
  • Steven W. Zucker

Abstract

We introduce a novel notion of scale for signals which we illustrate for shape. It is based on the notion of entropy and a view of shocks as “black holes of information”. We propose that to properly place features in a hierarchy, we need both linear, global, and instantaneously propagated smoothing (e.g. Gaussian smoothing), as well as non-linear, local smoothing (e.g., certain morphological operators) that propagate information with finite speed. Both of these types of processes are brought together in the entropy scale space. The scheme is illustrated for shape, is intuitive and robust, and applicable in other areas of vision.5

Keywords

Black Hole Rarefaction Wave Scale Space Jump Condition Morphological Operation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    J. Babaud, A. P. Witkin, M. Baudin, and R. O. Duda. Uniqueness of the gaussian kernel for scale-space filtering. PAMI, 8(1):26–33, January 1986.MATHCrossRefGoogle Scholar
  2. [2]
    F. W. Campbell and J. G. Robson. Applications of fourier analysis to the visibility of gratings. J. Physiol. (London), 197:551–556, 1968.Google Scholar
  3. [3]
    M.-H. Chen and P.-F. Yan. A multiscale approach based on morphological filtering. PAMI, 11(7):694–700, July 1989.MATHCrossRefGoogle Scholar
  4. [4]
    P. Garabedian. Partial Differential Equations. John Wiley and Sons, 1964.Google Scholar
  5. [5]
    E. Hopf. The partial differential equation u t + uu x = eu x x. Comm. Pure Appl. Math., 3:201–230, 1950.MathSciNetMATHCrossRefGoogle Scholar
  6. [6]
    R. Hummel, B. B. Kimia, and S. W. Zucker. Deblurring gaussian blur. Computer Vision, Graphics, and Image Processing, 38:66–80, 1987.MATHCrossRefGoogle Scholar
  7. [7]
    B. B. Kimia. Conservation laws and a theory of shape. Ph.D. dissertation, McGill Centre for Intelligent Machines, McGill University, Montreal, Canada, 1990.Google Scholar
  8. [8]
    B. B. Kimia, A. Tannenbaum, and S. W. Zucker. On the evolution of curves via a function of curvature, I: The classical case. JMAA, To appear, 1991.Google Scholar
  9. [9]
    J. J. Koenderink. The structure of images. Biol. Cybernetics, 50:363–370, 1984.MathSciNetMATHCrossRefGoogle Scholar
  10. [10]
    J. J. Koenderink and A. J. van Dooran. Visual detection of spatial contrast: Influence of location in the visual field, target extent and illuminance level. Biological Cybernetics, 30:157–167, 1978.CrossRefGoogle Scholar
  11. [11]
    J. J. Koenderink and A. J. van Doom. Dynamic shape. Biol. Cyber., 53:383–396, 1986.MATHCrossRefGoogle Scholar
  12. [12]
    P. D. Lax. Shock Waves and Entropy, pages 603–634. Academic Press, New York, 1971.Google Scholar
  13. [13]
    P. D. Lax. Hyperbolic Systems of Conservation Laws and the Mathematical Theory of Shock Waves. SIAM Regional Conference series in Applied Mathematics, Philadelphia, 1973.Google Scholar
  14. [14]
    P. D. Lax. Shock waves, increase of entropy and loss of information. Technical Report DOE/ER/03077-240, Courant Mathematics and Computing Library-R&D Report, 1984.Google Scholar
  15. [15]
    F. Leymarie and M. D. Levine. Shape features using curvature morphology. In Proceedings of the SPIE “Intelligent Robots and Computer Vision VIII: Algorithms and Techniques” Conference, volume 1192, part 2, pages 536-547, Philadelphia, Penn., USA, November 1989.Google Scholar
  16. [16]
    P. Maragos. Pattern spectrum and multiscale shape representation. PAMI, 11(7):701–716, July 1989.MATHCrossRefGoogle Scholar
  17. [17]
    D. Marr and E. Hildreth. Theory of edge detection. Technical Report MIT AI Memo 518, MIT AI Lab, 1979.Google Scholar
  18. [18]
    F. Mokhtarian and A. Mackworth. Scale-based description of planar curves and two-dimensional shapes. PAMI, 8:34–43, 1986.CrossRefGoogle Scholar
  19. [19]
    O. Oleinik. Discontinuous solutions of nonlinear differential equations. Amer. Math. Soc. Transl. Ser. 2, 26:95–172, 1957.Google Scholar
  20. [20]
    P. Perona and J. Malik. Scale-space and edge detection using anistroopic diffusion. PAMI, 12(7):629–639, July 1990.CrossRefGoogle Scholar
  21. [21]
    W. Richards, B. Benjamin, and D. Whittington. Encoding contour shape by curvature extrema. JOSA A, 3(9):1483–1489, 1986.CrossRefGoogle Scholar
  22. [22]
    A. Rosenfeld and M. Thurston. Edge and curve detection for visual scene analysis. IEEE Trans. Comput., C-20:562–569, 1971.CrossRefGoogle Scholar
  23. [23]
    J. Smoller. Shock Waves and Reaction-Diffusion Equations. Springer-Verlag, New York, 1983.MATHCrossRefGoogle Scholar
  24. [24]
    D. Widder. The Heat Equation. Academic Press, 1975.Google Scholar
  25. [25]
    H. R. Wilson and J. R. Bergen. A four mechanism model for threshold spatial vision. Vision Research, 19(1):19–32, 1979.CrossRefGoogle Scholar
  26. [26]
    A. P. Witkin. Scale-space filtering. In Proceedings of the 8 th International Joint Conference on Artificial Intelligence, pages 1019-1022, Karlsruhe, West Germany, Aug. 1983.Google Scholar
  27. [27]
    A. L. Yuille and T. A. Poggio. Scaling theorems for zero crossings. PAMI, 8(1):15–25, January 1986.MATHCrossRefGoogle Scholar
  28. [28]
    S. W. Zucker and R. Hummel. Receptive fields and the representation of visual information. Human Neurobiology, 5:121–128, 1986.Google Scholar

Copyright information

© Springer Science+Business Media New York 1992

Authors and Affiliations

  • Benjamin B. Kimia
    • 1
  • Allen Tannenbaum
    • 1
    • 2
    • 3
  • Steven W. Zucker
    • 1
    • 4
    • 5
  1. 1.Laboratory for Engineering Man-Machine SystemsBrown UniversityProvidenceUSA
  2. 2.Dept. of Elect. Eng.University of MinnesotaMinneapolisUSA
  3. 3.Department of Electrical EngineeringTechnionCity-HaifaIsrael
  4. 4.Canadian Institute for Advanced ResearchCanada
  5. 5.McGill Research Center For Intelligent MachinesCanada

Personalised recommendations