Thai Character Recognition Using “Snakecut” Technique

  • Thakerng Wongsirichot
  • Piyawan Seekaew
  • Parunyu Arnon
Conference paper
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 95)


Optical character recognition has become one of the powerful tools in the area of pattern recognition and artificial intelligence. Our research targets practically to Thai character recognition on Thai national identification cards. We propose a Thai character recognition technique, so called “Snakecut”. It is invented by iterative investigations of the Thai characters. The “Snakecut”, an unsupervised technique, is able to trace around each character, meticulously. A set of 26 weighted parameters are employed to further tuning recognition performances. Experiment results accuracy rates are 83.10% for pure Thai characters and 92.96% for a mixture of numbers and Thai characters with some limitations.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Alata, M., Al-shabi, M.: Text detection and character recognition using fuzzy image processing. Journal of Electrical Engineering 57, 258–267 (2006)Google Scholar
  2. 2.
    Baeza-Yates, R., Ribeiro-Neto, B.: Modern Information Retrieval. Addison Wesley, Reading (1999)Google Scholar
  3. 3.
    Hara, S.: Ocr for japanese classical documents segmentation of cursive characters. In: PNC 2004 Annual Conference on Conjunction with PRDLA, National Institute of Japanese (2004)Google Scholar
  4. 4.
    Kawtrakul, A., Waewsawangwong, P.: Multi-feature extraction for printed thai character recognition. In: Fourth Symposium on Natural Language Processing, pp. 1–8 (2000)Google Scholar
  5. 5.
    Krstovski, K., MacRostie, E., Prasad, R., Natarajan, P.: End-to-end trainable thai ocr system using hidden markov models. In: IAPR International Workshop on Document Analysis Systems, vol. 0, pp. 607–614 (2008)Google Scholar
  6. 6.
    Mitrpanont, J., Kiwprasopsak, S.: The development of the feature extraction algorithms for thai handwritten character recognition system. In: Hendtlass, T., Ali, M. (eds.) IEA/AIE 2002. LNCS (LNAI), vol. 2358, pp. 77–89. Springer, Heidelberg (2002)CrossRefGoogle Scholar
  7. 7.
    Tanprasert, C., Sae-Tang, S.: Thai type style recognition. In: IEEE International Symposium on Circuit and Systems, pp. 336–339 (1999)Google Scholar
  8. 8.
    The Ministry of Interior Department of Provincial Administration, Thai national identification cardsGoogle Scholar
  9. 9.
    Wyatt, D.K.: A Short History of Thailand. Yale University Press, New Haven (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Thakerng Wongsirichot
    • 1
  • Piyawan Seekaew
    • 1
  • Parunyu Arnon
    • 1
  1. 1.Information and Communication Technology Programme, Faculty of SciencePrince of Songkla UniversitySongklaThailand

Personalised recommendations