A Recognition System for Multi-Form Korean Characters Based on Hierarchical Temporal Memory

  • Haibao, Nan (Department of Computer Science, GyeongSang National University) ;
  • Bae, Sun-Gap (Department of Computer Science, GyeongSang National University) ;
  • Bae, Jong-Min (Department of Computer Science, GyeongSang National University) ;
  • Kang, Hyun-Syug (Department of Computer Science, GyeongSang National University)
  • Published : 2009.12.30

Abstract

Traditional character recognition systems usually aim at characters with simple variation. With the development of multimedia technology, printed characters may appear more diversely. Existing recognition technologies can't deal with Hangul recognition effectively in diverse environments. This paper presents a recognition system for multi-form Korean characters called RSMFK, which is based on the model of Hierarchical Temporal Memory (HTM). Our system can effectively recognize the printed Korean characters of different fonts, scales, rotation, noise and background. HTM is a model which simulates the neocortex of human brain to recognize and memorize intelligently. Experimental results show that RSMFK performs a good recognition rate of 97.8% on average, which is proved to be obviously improved over the conventional methods.

Keywords

References

  1. S. Araokar, "Visual Character Recognition using Artificial Neural Networks," Computer Science Neural and Evolutionary Computing, 2005.
  2. K.J.Pyeoung and H.B.Jeen, "Hangul Recognition on Multi-Transputer System [J]," IEEE Region 10 Conference on Communication, Control and Power Engineering, Vol.2, pp. 1102-1105, 1993.
  3. K. Kang and J. Kim, "Handwritten Hangul Character Recognition with Hierarchical Stochastic Character Representation [A]," IEEE Computer Society, pp. 212-216 Vol.1, 2003.
  4. P. Lee and H. Jang, "A study on printed Hangul recognition with dynamic Jaso segmentation and neural network [J]," Journal of Korean Institute of Communication, pp. 2133-2145, 1994(19).11.
  5. J. koh, G. S. Moon, K. G. Mehrota, C. K. Mohan and S. Ranka, "Korean Character Recognition Using Neural Networks [J]," 3rd Symposium on the Frontiers of Massively Parallel Computation, pp. 162-165, 1990.
  6. K. Ham, "A Study for Printed Korean Character Segmentation and Recognition by Vowel Pre Recognition," Paper of Conference of Dongguk University, Vol.10, pp. 555-567, 1991.
  7. J. Kim, K. Kim and S. Chien, "Korean and English Character Recognition System Using Hierarchical classification Neural Network," IEEE International Conference on Digital Object Identifier, pp. 759-764 Vol.1, 1995.
  8. J. Hawkins and S. Blakeslee, On Intelligence, New York, Henry Holt, 2004. http://www.onintelligence.com
  9. D. George and B. Jaros. "The HTM Learning Algorithms," Numenta Inc., 2007. http://www.numenta.com/for-developers/education/Numenta_HTM_Learning_Algos.pdf
  10. R.Xu and C.P.Liu, "Grapheme Segmentation and Recognition in Machine Printed Hangul Characters," Journal of Chinese Information Processing, pp. 66-71, 2006(26).
  11. Numenta Inc., "NuPIC Getting Started," http://www.numenta.com/
  12. Vitamin D Inc., "Vitamin D Toolkit Reference Guide," http://www.numenta.com/