Visual Form pp 421-430 | Cite as

Shape Classification by Using Associative Memories

  • Giancarlo Parodi
  • Gianni Vernazza
  • Rodolfo Zunino

Abstract

The use of associative techniques to perform shape recognition is considered. With respect to classical recognition techniques, it is shown how the pattern-completion capability and content addressability of associative memories can yield robust performance and high computational speed in shape recognition applications, especially in those domains in which promptness of ‘first-glance’ classification is more important than accurate pattern-analysis capabilities. The noise-like coding model of associative memory is adopted, specifying the basic associative classification principle. A set of different experimental domains are considered, at increasing levels of complexity, evidencing how the theoretical model can be effectively applied to image and shape recognition.

Keywords

Associative Memory Fourier Descriptor Shape Recognition Associative Model Input Shape 
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]
    Bottini S (1980) An algebraic model of an associative noise-like coding memory. Biol Cybern 36:221–228.MATHCrossRefGoogle Scholar
  2. [2]
    Bottini S (1988) An after-Shannon measure of the storage capacity of an associative noise-like coding memory. Biol Cybern 59:151–159.MathSciNetCrossRefGoogle Scholar
  3. [3]
    Parodi GC, Zunino R (1990) Image classification by using associative noise-like memories. Proc. IEEE Int.Phoenix Conf.on Comp.and Comm. IPCCC’90 Phoenix AZ March 1990.Google Scholar
  4. [4]
    Parodi GC, Zunino R (1990) Noise insensitivity of an associative image classification system. Proc. IEEE Int.Workshop on Robust Comp.Vision, Seattle WA — in press by Academic Press New York.Google Scholar
  5. [5]
    Parodi GC, Regazzoni C, Vernazza G, Zunino R (1991) Speeding up scene recognition by using an associative noise-like coding memory. IEEE Int.Phoenix Conf. on Comp.and Comm. IPCCC’91 Phoenix AZ March 1991.Google Scholar
  6. [6]
    Persoon E, Fu KS (1977) Shape discrimination using Fourier descriptors. IEEE Trans on Sys Man Cybern vol. SMC-7, pp. 170–179.MathSciNetCrossRefGoogle Scholar
  7. [7]
    Arbter K, Snyder WE, Burkhardt H, Hirzinger G, (1990) Application of affine-in-variant Fourier descriptors to recognition of 3-D objects. IEEE Trans Pattern Anal. Machine Intell July 1990, vol. PAMI-12 no. 7 pp. 640–647.CrossRefGoogle Scholar
  8. [8]
    Kie-Bum Eom, Juha Park (1990) A robust model based approach in shape recognition. Proc. IEEE Int.Workshop Robust Comp.Vision, Seattle Oct 1990, pp. 57–71 — in press by Academic Press, New York.Google Scholar
  9. [9]
    Chen DS, Schunck BG (1990) Robust statistical methods for building classification procedures. Proc. IEEE Int.Workshop Robust Comp.Vision, Seattle Oct 1990, pp. 72–85 — in press by Academic Press, New York.Google Scholar
  10. [10]
    Borsellino A, Poggio T (1973) Convolution and correlation algebras. Kybernetik 10:113–122.MathSciNetCrossRefGoogle Scholar
  11. [11]
    Hinton GE, Anderson JA (Eds) (1989) Parallel models of associative memory (Updated Edition). Lawrence Erlbaum Ass., Hillsdale, NJ 1989.MATHGoogle Scholar
  12. [12]
    Wechsler H, Zimmermann JL (1988) 2-D invariant object recognition using distributed associative memory. IEEE Trans PAMI Vol 10-6: 811–821.CrossRefGoogle Scholar
  13. [13]
    Polzleitner W, Wechsler H (1990) Selective and Focused invariant recognition using distributed associative memories (DAM). IEEE Trans PAMI Vol. 12-8: 809–814.CrossRefGoogle Scholar
  14. [14]
    Fukushima K, Miyake S, Ito T (1983) Neocognitron: a neural network model for a mechanism of visual pattern recognition. IEEE Trans Sys Man and Cybern Sept/Oct 1983: 826-834.Google Scholar
  15. [15]
    Rumelhart DE, McClelland JL (1986) Parallel distributed processing — Vols. 1 and 2. MIT Press 1986.Google Scholar
  16. [16]
    Carpenter GA, Grossberg S (1987) A massively parallel architecture for a self-organizing neural pattern recognition machine. Comp. Vision, Graphics, and Image Processing 1987, 37:54–115.MATHCrossRefGoogle Scholar
  17. [17]
    Zwicke PE, Kiss I Jr (1983) A new implementation of the Mellin transform and its application to radar classification of ships. IEEE Trans PAMI Vol 5-2:191–199.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 1992

Authors and Affiliations

  • Giancarlo Parodi
    • 1
  • Gianni Vernazza
    • 1
  • Rodolfo Zunino
    • 1
  1. 1.Department of Biophysical and Electronic Engineering (DIBE)University of GenoaGenoaItaly

Personalised recommendations