Ancient Document Recognition Using Fuzzy Methods

Cláudia S. Ribeiro, João M. Gil, João R. Caldas Pinto, João M. Sousa

Abstract

This paper proposes an optical character recognition system based on fuzzy logic for 17th century printed documents. The process consists of two stages: training with collected character image examples and new character image classification. Training builds fuzzy membership functions from aligned oriented features extracted using Gabor filters. These are used in classification to select a most likely character group for new data. A post-processing stage with a proposed modified Levenshtein word distance metric further improves results. A success rate of 88% is achieved on a significant test set.

References

  1. R. Buse, Z.Q. Liu, J. Bezdek, “Word Recognition Using Fuzzy Logic”, in IEEE Transactions on Fuzzy Systems, vol. 10, no. 1, Fev. 2001, pp. 65-76
  2. João M.C. Sousa and Uzay Kaymak, "Fuzzy Decision Making in Modeling and Control", World Scientific, Singapore and UK, Dec. 2002
  3. R. Buse, Z.Q. Liu, T. Caelli, “A structural and relational approach to handwritten word recognition”, in IEEE Trans. Syst., Man, Cybern., vol. 27, no. 25, Oct. 1997, pp 847-861
  4. Parker, J.R., “Algorithms for Image Processing and Computer Vision”, John Wiley & Sons, New York, USA, 1998
  5. N. Otsu, "A threshold selection method from gray level histograms", IEEE Transactions on Systems, Man, and Cybernetics, vol. 9, pp. 62-66, 1979
  6. Godfried Toussaint, “Solving Geometric Problems with the Rotating Calipers”, in Proc. IEEE MELECON'83, 1983, pp. A10.02/1-4
  7. James D. Foley et al, “Computer Graphics - Principles and Practice”, Second Edition in C, Addison-Wesley, Reading, Massachussets, USA, 1990
  8. C. L. Hwang, K. Yoon, “Multiple Attribute Decision Making, Methods and Applications, A State-of-the-Art-Survey”, Springer-Verlag, Berlin, Germany, 1981
  9. K. Atkinson, GNU Aspell Homepage, http://aspell.net, GNU Project
  10. V. I. Levenshtein, "Binary codes capable of correcting deletions, insertions and reversals", Soviet Physics Doklady, vol. 6, pp. 707-710, 1966
  11. Álvaro Ferreira de Véra, “Orthographia ou modo para escrever certo na lingua Portuguesa”, 17th century, available at Biblioteca Nacional
  12. ABBYY FineReader Homepage, http://www.abbyy.com, ABBYY Software House
Download


Paper Citation


in Harvard Style

S. Ribeiro C., M. Gil J., R. Caldas Pinto J. and M. Sousa J. (2004). Ancient Document Recognition Using Fuzzy Methods . In Proceedings of the 4th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2004) ISBN 972-8865-01-5, pages 98-107. DOI: 10.5220/0002685600980107


in Bibtex Style

@conference{pris04,
author={Cláudia S. Ribeiro and João M. Gil and João R. Caldas Pinto and João M. Sousa},
title={Ancient Document Recognition Using Fuzzy Methods},
booktitle={Proceedings of the 4th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2004)},
year={2004},
pages={98-107},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0002685600980107},
isbn={972-8865-01-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 4th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2004)
TI - Ancient Document Recognition Using Fuzzy Methods
SN - 972-8865-01-5
AU - S. Ribeiro C.
AU - M. Gil J.
AU - R. Caldas Pinto J.
AU - M. Sousa J.
PY - 2004
SP - 98
EP - 107
DO - 10.5220/0002685600980107