Memory Networks for Practical Vision Systems: Design Calculations

  • I. Aleksander


If memory networks are to be used in real robotic applications, the designer needs some basis from which to calculate the size of the modules required and the amount of training that provides optimal results. This paper is concerned with the recognition of solid shapes, and provides design calculations and methods for systems that recognize objects irrespective of their location or orientation. It tackles the problem of the recognition of partially hidden or overlapping objects, and the measurement of position and orientation itself.


Finite State Machine Random Access Memory Image Recognition Training Pattern Artificial Vision 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    McCulloch, W. S.; Pitts, W. H. A logical calculus of the ideas imminent in nervous activity. Bulletin of Mathematical Biophysics 1943, 5, 15–33.CrossRefGoogle Scholar
  2. 2.
    Minsky, M. L.; Papert, S. Perceptrons: An Introduction to Computational Geometry MIT Press, Cambridge, 1969.Google Scholar
  3. 3.
    F. Principles of Neurodynamics Spartan, Washington, 1962.Google Scholar
  4. 4.
    Aleksander, I.; Albrow, R. C. Pattern recognition with adaptive logic circuits. Proceedings of IEE—NPL Conference on Pattern Recognition 1968, 42, 1–10.Google Scholar
  5. 5.
    Aleksander, I., Stonham, T. J., Wilkie, B. A. Computer vision systems for industry. Digital Systems for Industrial Automation 1982, 1 (4).Google Scholar
  6. 6.
    Widrow, B. Generalisation and information storage in networks of adaptive neurons. In Self-organising Systems eds Yovits et al; Spartan, New York, 1962.Google Scholar
  7. 7.
    Wilkie, B. A. Design of a High-resolution Adaptive Pattern RecogniserThesis, Brunel University, in preparation.Google Scholar
  8. 8.
    British Patent Application No. 8135939 November, 1981.Google Scholar
  9. 9.
    Bledsoe, W. W.; Browning, I. Pattern recognition and reading by machine. Proceedings East. J.C.C. 1959, 225–232.Google Scholar
  10. 10.
    Ullman, J. R. Experiments with the N-tuple method of pattern recognition. IEEE Transactions on Computers 1969, 1135.Google Scholar
  11. 11.
    Aleksander, I.; Albrow, R. C. Microcircuit learning nets: hamming distance behaviour. Electronics Letters 1970, 6, 134–135.CrossRefGoogle Scholar
  12. 12.
    Aleksander, I.; Stonham, T. J. A guide to pattern recognition using random access memories. IEE Proceedings on Computers and Digital Techniques 1979, 2, 29–36.CrossRefGoogle Scholar
  13. 13.
    Dessimoz, J. D., Kammennos, P. Software for robot vision. Digital Systems for Industrial Automation 1982, 1, 143–160.Google Scholar
  14. 14.
    Reeves, A. P. Tracking Experiments with an Adaptive Logic System Thesis, University of Kent, 1973.Google Scholar
  15. 15.
    Dawson, C. Simple Scene Analysis Using Digital Learning Nets Thesis, University of Kent, 1975.Google Scholar
  16. 16.
    Fairhurst, M. C. Some Aspects of Learning in Digital Adaptive Networks Thesis, University of Kent, 1973.Google Scholar
  17. 17.
    Tolleyfield, A. J. Some Properties of Sequential Learning Networks Thesis, University of Kent, 1975.Google Scholar
  18. 18.
    Aleksander, I. Action-orientated learning networks. Kybernetes 1975, 4, 39–44.CrossRefGoogle Scholar
  19. 19.
    Wilson, M. J. D. Artificial perception in adaptive array. IEEE Transactions on Systems, Man and Cybernetics 1980, 10, 25–32.CrossRefGoogle Scholar

Copyright information

© Igor Aleksander 1983

Authors and Affiliations

  • I. Aleksander
    • 1
  1. 1.Department of Electrical Engineering and ElectronicsBrunel UniversityEngland

Personalised recommendations