A RECURRENT NEURAL NETWORK RECOGNISER FOR ONLINE RECOGNITION OF HANDWRITTEN SYMBOLS

Bing Quan Huang, Tahar Kechadi

2005

Abstract

This paper presents an innovative hybrid approach for online recognition of handwritten symbols. This approach is composed of two main techniques. The first technique, based on fuzzy logic, deals with feature extraction from a handwritten stroke and the second technique, a recurrent neural network, uses the features as an input to recognise the symbol. In this paper we mainly focuss our study on the second technique. We proposed a new recurrent neural network architecture associated with an efficient learning algorithm. We describe the network and explain the relationship between the network and the Markov chains. Finally, we implemented the approach and tested it using benchmark datasets extracted from the Unipen database.

References

  1. B.Q. Huang, T. R. and Kechadi, T. (2004). A new modifed network based on the elman network. In Proc. of IASTED Internation Conferance on Artificial Intellegence and Application, Innsbruck, Austria.
  2. C.C.Tappet (June 1984). Adaptive on-line handwriting recognition. In Proc. 7th Int. conf. on Pattern Recognition, pages 1004-1007, Montreal, Canada.
  3. Elman, J. (1999). Finding structure in time. Cognitive Science, 14(2):179-211.
  4. Fausett, L. (1994). Fundamentals of Neural Networks. Englewood Cliffs, NJ: Prentice Hall.
  5. Gomes, N. and Ling, L. L. (2001). Feature extraction based on fuzzy set theory for handwriting recognition. In ICDAR'01, pages 655-659.
  6. Guyon, I., S. L. P. R. L. M. and Janet, S. (Oct, 1994). Unipen project of on-line data exchange and recognizer benchmarks. Proc. of the 12th International Conference on Pattern Recognition,ICPR'94, pages 29-33.
  7. J.A.Fitzgerald, F. and T.Kechadi (May 2004). Feature extraction of handwritten symbols using fuzzy logic. In The Seventeenth Canadian Conference on Artificial Intelligence, pages 493-498.
  8. J.Hu, M. and W.Turin (Oct, 1996). Hmm based on-line handwriting recognition. IEEE Trans. Pattern Analysis and Machine Intelligence, 18(10):1039-1045.
  9. Lawrence, S., Giles, C. L., and Fong, S. (2000). Natural language grammatical inference with recurrent neural networks. IEEE Trans. on Knowledge and Data Engineering, 12(1):126-140.
  10. L.A.Zadeh (1972). Outline of a new approach to the analysis of complex systems and decision processes. In Man and Computer, pages 130-165.
  11. L.R. Bahl, F. and Mercer, R. (March 1983). A maximum likelihood approach to continuous speech recognition. IEEE Trans. Pattern Analysis and Machine Intelligence, 5(3):179-190.
  12. M. Schnekel, I. G. and Henderson, D. (April 1994). On-line cursive script recognition using time delay networks and hidden markove models. In Proc. ICASSP'94, volume 2, pages 637-640, Adelaide,Australia.
  13. M.A. Castao, F. C. (1997). Training simple recurrent networks through gradient descent algorithms. In Lecture Notes in Comp. Sci.: Biological & Artificial Computation: From Neuroscience to Technology, volume 1240, pages 493-500. Springer Verlag.
  14. Malaviya, A. and Peters, L. (1997). Fuzzy feature description of handwriting patterns. Pattern Recognition, 30(10):1591-1604.
  15. O. D. Trier, A. K. J. and Taxt, T. (1996). Feature extraction methods for character recognition - a survey. In Pattern Recognition, pages 641-662.
  16. Plamondon, R. and Maarse, F. J. (May 1989). An evaluation of motor models of handwriting. IEEE Trans. systems Man, and Cybernetics, 19:1060-1072.
  17. Rabiner, L. (Feb 1989). A tutorial on hidden markov models and selected applications in speech recognition. Proc. of the IEEE, 77(2).
  18. R.J.Williams and Zipser, D. (1995). Gradient-based learning algorithms for recurrent networks and their computational complexity. In Chauvin, Y. and Rumelhart, D. E., editors, Back-propagation: Theory, Architectures and Applications, chapter 13, pages 433-486. Lawrence Erlbaum Publishers, Hillsdale, N.J.
  19. S. Bercu, G. L. (May 1993). On-line handwritten word recognition: An approach based on hidden markov models. In Proc. 3r Int.Workshop on frontiers in Handwriting Recognition, pages 385-390, buffalo, New York.
  20. Schomaker, L. R. B. and Tteulings, H.-L. (April 1990). A handwriting recognition system based on the properties and architectures of the human motor system. In Proc. Int. Workshop on frontiers in Handwriting Recognition, Montreal, pages 195-211.
  21. S.E. Levinson, L. R. and Sondhi, M. (April 1983). An into the application of the theory of probabilistic functions of a markov process to automatic speech recognition. Bell System Technical Joural, 62(4):1035-1074.
  22. Seni, G. and Nasrabadi, N. (1994). An on-line cursive word recognition system. In CVPR94, pages 404-410.
  23. Subrahmonia, J. and Zimmerman, T. (2000). Pen computing: Challenges and applications. In Proc. ICPR 2000, pages 2060-2066.
  24. T.Wakahara and K.Odaka (Dec. 1997). On-line cursive kanji character recognition using stroke-based afine transformation. IEEE Trans. Pattern Analysis and Machine Intelligence, 19(12).
  25. Werbos, P. J. (1990). Backpropagation through time: what it does and how to do it. In Proc. of the IEEE, volume 78, pages 1550-1560.
  26. Williams, R. and Zipser, D. (1989). A learning algorithm for continually running fully recurrent neural networks. Neural Computation, 1(2):270-280.
  27. Wilson, W. H. (1996). Learning performance of networks like elman's simple recurrent netwroks but having multiple state vectors. Workshop of the 7th Australian Conference on Neural Networks, Australian National University Canberra.
  28. Zadeh, L. (1975). Calculus of fuzzy restrictions. In Fuzzy Sets and Their Applications to Cognitive and Decision Processes, pages 1-39, Academic Press, NY.
Download


Paper Citation


in Harvard Style

Quan Huang B. and Kechadi T. (2005). A RECURRENT NEURAL NETWORK RECOGNISER FOR ONLINE RECOGNITION OF HANDWRITTEN SYMBOLS . In Proceedings of the Seventh International Conference on Enterprise Information Systems - Volume 2: ICEIS, ISBN 972-8865-19-8, pages 27-34. DOI: 10.5220/0002514200270034


in Bibtex Style

@conference{iceis05,
author={Bing Quan Huang and Tahar Kechadi},
title={A RECURRENT NEURAL NETWORK RECOGNISER FOR ONLINE RECOGNITION OF HANDWRITTEN SYMBOLS},
booktitle={Proceedings of the Seventh International Conference on Enterprise Information Systems - Volume 2: ICEIS,},
year={2005},
pages={27-34},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002514200270034},
isbn={972-8865-19-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Seventh International Conference on Enterprise Information Systems - Volume 2: ICEIS,
TI - A RECURRENT NEURAL NETWORK RECOGNISER FOR ONLINE RECOGNITION OF HANDWRITTEN SYMBOLS
SN - 972-8865-19-8
AU - Quan Huang B.
AU - Kechadi T.
PY - 2005
SP - 27
EP - 34
DO - 10.5220/0002514200270034