Using Contextual Information to Selectively Adjust Preprocessing Parameters

  • Predrag Neskovic
  • Leon N. Cooper
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2085)

Abstract

It is generally accepted that some of the problems and ambiguities at the low level of recognition and processing can not be resolved without taking into account contextual expectations. However, in most recognition systems in use today, preprocessing is done using only bottom-up information. In this work we present a working system that is inspired by human perception and can use contextual information to modify preprocessing of local regions of the input pattern in order to improve recognition. This is especially useful, and often necessary, during the recognition of complex objects where changing the preprocessing of one section improves recognition of that section but has an adverse effect on the rest of the object. We present the results of our recognition system applied to on-line cursive script where in some cases the error rate is decreased by 20%.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Y. Bengio, Y. LeCun, C. Nohl, and C. Burges. Lerec: A NN/HMM hybrid for on-line handwriting recognition. Neural Computation, 7:1289–1303, 1995.CrossRefGoogle Scholar
  2. 2.
    H. Bourlard and C. Wellekens. Links between hidden Markov models and multilayer perceptrons. IEEE Transactions on Pattern Analysis and Machine Intelligence, 12:1167–1178, 1990.CrossRefGoogle Scholar
  3. 3.
    M. Cote, E. Lecolinet, M. Cheriet, and C. Suen. Automatic reading of cursive scripts using a reading model and perceptual concepts. Internatinal Journal of Document Analysis and Recognition, 1997.Google Scholar
  4. 4.
    W. Guerfali and R. Plamondon. Normalizing and restoring on-line handwriting. Pattern Recognition, 26(3):419–431, 1993.CrossRefGoogle Scholar
  5. 5.
    J. Hu, M. Brown, and W. Turin. HMM based on-line handwriting recognition. IEEE PAMI, 18(10):1039–1045, 1996.Google Scholar
  6. 6.
    J. McClelland and D. Rumelhart. An interactive activation model of context effects in letter perception. Psychological Reviews, 88:375–407, 1981.CrossRefGoogle Scholar
  7. 7.
    P. Morasso, L. Barberis, S. Pagliano, and D. Vergano. Recognition experiments of cursive dynamic handwriting with self-organizing networks. Pattern Recognition, 26(3):451–460, 1993.CrossRefGoogle Scholar
  8. 8.
    K. Nathan, H. Beigi, J. Subrahmonia, G. Clary, and H. Maruyama. Real-time on-line unconstrained handwriting recognition using statistical methods. In International Conference on Acoustics, Speech and Signal Processing, volume 4, pages 2619–2613, 1995.Google Scholar
  9. 9.
    P. Neskovic and L. Cooper. Neural network-based context driven recognition of on-line cursive script. In 7th International Workshop on Frontiers in Handwriting Recognition, pages 352–362, 2000.Google Scholar
  10. 10.
    P. Neskovic, P. Davis, and L. Cooper. Interactive parts model: an application to recognition of on-line cursive script. In Advances in Neural Information Processing Systems, 2000.Google Scholar
  11. 11.
    D.E. Rumelhart. Theory to practice: A case study-recognizing cursive handwriting. In E.B. Baum, editor, Computational Learning and Cognition: Proceedings of the Third NEC Research Symposium. SIAM, Philadelphia, 1993.Google Scholar
  12. 12.
    M. Schenkel, I. Guyon, and D. Henderson. On-line cursive script recognition using time delay neural networks and hidden markov models. Machine Vision and Applications, 8:215–223, 1995.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Predrag Neskovic
    • 1
  • Leon N. Cooper
    • 1
  1. 1.Physics Department and Institute for Brain and Neural SystemsBrown UniversityProvidenceUSA

Personalised recommendations