Advertisement

Segmentation Strategy of Handwritten Connected Digits (SSHCD)

  • Abdeldjalil Gattal
  • Youcef Chibani
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6979)

Abstract

The handwritten digit segmentation is the most important module for handwritten digit recognition, which constitutes a difficult task because of overlapping and / or connected of adjacent digits. To resolve this problem, several segmentation methods have been developed each one having its advantage and disadvantage. In this work, we propose a segmentation approach depending of the configuration link between digits. With the help of a few rules, multiple hypotheses are defined for finding the best segmentation path in order to separate two connected digits. Hence, a verification strategy is proposed in order to generate all possible segmentation-recognition hypotheses. The performance of our strategy is evaluated in terms of correct recognition rates using the confusion matrix.

Keywords

recognition segmentation segmentation-recognition handwritten digits verification strategy Support Vector Machines 

References

  1. 1.
    Dimauro, G., Impedovo, S., Pirlo, G., Salzo, A.: Automatic Bankcheck processing: A New Engineered System. International Journal of Pattern Recognition and Artificial Intelligence 11(4), 467–504 (1997)CrossRefGoogle Scholar
  2. 2.
    Vellasques, E., Oliveira, L.S., Britto Jr., A.S., Koerich, A.L., Sabourin, R.: Filtering segmentation cuts for digit string recognition. Pattern Recognition 41(10), 3044–3053 (2008)CrossRefzbMATHGoogle Scholar
  3. 3.
    Congedo, G., Dimauro, G., Impedovo, S., Pirlo, G.: Segmentation of Numeric Strings. In: Proc. of Third Int. Conf. on Document Analysis and Recognition, Canada, pp. 1038–1041. IEEE Computer Society, Montreal (1995)Google Scholar
  4. 4.
    Jang, B.K., Chin, R.T.: One-pass parallel thinning: Analysis, properties, and quantitative evaluation. IEEE Transactions on Pattern Analysis and Machine Intelligence 14(11), 1129–1140 (1992)CrossRefGoogle Scholar
  5. 5.
    Shridhar, M., Badreldin, A.: Recognition of Isolated and Simply Connected Handwritten Numerals. Journal of Pattern Recognition 19(1), 1–12 (1986)CrossRefGoogle Scholar
  6. 6.
    Ayat, N.E., Cheriet, M., Suen, C.Y.: Un système neuro-flou pour la reconnaissance de montants numériques de chèques arabes. Colloque international francophone sur l’écrit et le document, Montréal, Québec, Canada, pp. 03–07 (2000)Google Scholar
  7. 7.
    Hussein, K.M., Agarwal, A., Gupta, A., Wang, P.S.P.: A knowledge-based algorithm for enhanced recognition of handwritten courtesy amounts. Pattern Recognition 32, 305–316 (1999)CrossRefGoogle Scholar
  8. 8.
    Oliveira, L.S., Sabourin, R., Bortolozzi, F., Suen, C.Y.: A modular system to recognize numerical amounts on Brazilian bank cheques. In: Proc. of 6th International Conference on Document Analysis and Recognition (ICDAR), Seattle, USA, pp. 389–394. IEEE Computer Society, Los Alamitos (2001)CrossRefGoogle Scholar
  9. 9.
    Fujisawa, H., Nakano, Y., Kurino, K.: Segmentation Methods for Character Recognition: From Segmentation to Document Structure Analysis. Proceedings of the IEEE 80(7), 21–28, 1079–1091 (1996)Google Scholar
  10. 10.
    Grother. P.J.: NIST Special Database 19; Handprinted Forms and Characters Database. National Institute of Standards and Technology, NIST (1995)Google Scholar
  11. 11.
    Schölkopf, B., Burges, C., Vapnik, V.: Extracting support data for a given task. In: KDD 1995, pp. 252–257 (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Abdeldjalil Gattal
    • 1
    • 2
  • Youcef Chibani
    • 3
  1. 1.Université de TébessaAlgeria
  2. 2.Ecole Nationale Supérieure d’Informatique (ESI)Oued SmarAlgeria
  3. 3.Laboratoire de Communication Parlée et Traitement des Signal, Faculté d’ Electronique et d’InformatiqueUniversity of Sciences and Technology Houari BoumedienneAlgiersAlgeria

Personalised recommendations