Segmentation-Free Strategy: Advanced Algorithms

  • Tonghua Su
Part of the SpringerBriefs in Electrical and Computer Engineering book series (BRIEFSELECTRIC)


The hidden Markov model (HMM), a powerful tool, has been widely applied to sequence analysis tasks such as speech recognition and handwriting recognition performance. However, its recognition performance is normally limited to the general framework. In this chapter, we investigate sophisticated techniques for improving its recognition performance. This includes a method for synthesizing string samples from isolated character images. Second, enhanced features are derived considering the uniqueness of Chinese characters and text line normalization is added to improve feature discrimination. Third, discriminative training based on MPE criteria is explored in the context of Chinese handwriting recognition for the first time. Fourth, a bridge is built between a segmentation-free and a segmentation-based system. This chapter also discusses the distributed training of the bigram language model.


Hide Markov Model Chinese Character Ensemble Method Text Line Gabor Feature 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. R. Bertolami, H. Bunke, Hidden Markov Model-based ensemble methods for offline handwritten text line recognition. Pattern Recogn. 41(11), 3346–3452 (2008)CrossRefGoogle Scholar
  2. Y. Chen, X. Ding, Y. Wu, Off-line handwritten Chinese character recognition based on crossing line feature. In: Proceedings of the Fourth International Conference on Document Analysis and Recognition, Ulm, Germany, 18–20 August 1997Google Scholar
  3. M. Cheriet, N. Kharma, C.-L. Liu, C.Y. Suen, Character Recognition Systems: A Guide for Students and Practitioners (Wiley, NJ, 2007)zbMATHCrossRefGoogle Scholar
  4. J. Dean , S. Ghemawat, MapReduce: simplified data processing on large clusters. In: Proceedings of the 6th conference on symposium on opearting systems design and implementation, USENIX Association, Berkeley, March 31, 2004Google Scholar
  5. J.E. Freund, Modern Elementary Statistics (Prentice-Hall, New Jersey, 1984)Google Scholar
  6. S. Furui, Speaker-independent isolated word recognition using dynamic features of speech spectrum. IEEE Trans ASSP 34(1), 52–59 (1986)CrossRefGoogle Scholar
  7. M. Gales, S. Young, The application of hidden Markov models in speech recognition. Foundations and Trends in Signal Processing 1(3), 195–304 (2007). doi: 10.1561/2000000004 zbMATHCrossRefGoogle Scholar
  8. Y. Ge, Q. Huo (2002) A comparative study of several modeling approaches for large vocabulary offline recognition of handwritten Chinese characters. In: International Conference on Pattern Recognition, IEEE Computer Society, Quebec, 11–15 August 2002, pp 85–88 Google Scholar
  9. H. Jiang, Discriminative training of HMMs for automatic speech recognition: A survey. Comput. Speech Lang. 24(4), 589–608 (2010)CrossRefGoogle Scholar
  10. C.-L. Liu, M. Koga, H. Fujisawa, Lexicon-driven segmentation and recognition of handwritten character strings for Japanese address reading. IEEE Trans. Pattern Anal. Mach. Intell. 24(11), 1425–1437 (2002)CrossRefGoogle Scholar
  11. U.V. Marti, H. Bunke, Use of positional information in sequence alignment for multiple classifier combination. In: International Workshop on Multiple Classifier Systems, Cambridge, England pp 388–398, 2001 Google Scholar
  12. M. Mohamed, P. Gader, Handwritten word recognition using segmentationfree hidden Markov modeling and segmentation-based dynamic programming techniques. IEEE. Trans. Pattern. Anal. Mach. Intell. 18, 548–554 (1996)CrossRefGoogle Scholar
  13. S. Natarajan, R. Saleem, E. Prasad, K. MacRostie, Subramanian Multi-lingual offline handwriting recognition using hidden Markov models: a script-independent approach. In: Lecture Notes in Computer Science, 2008, pp 231–250Google Scholar
  14. D. Povey, Discriminative training for large vocabulary speech recognition. PhD thesis, University of Cambridge, 2004Google Scholar
  15. T.H Su, C.L Liu, Improving HMM-based Chinese handwriting recognition using delta features and synthesized string samples. In: International Conference on Frontiers in Handwriting Recognition, 2010Google Scholar
  16. T.-H. Su, T.-W. Zhang, D.-J. Guan, Off-line Recognition of Realistic Chinese Handwriting Using Segmentation-free Strategy. Pattern Recogn. 42(1), 167–182 (2009)zbMATHCrossRefGoogle Scholar
  17. T-H. Su, T-W. Zhang, H-J. Huang, C-L Liu, Segmentation-free recognizer based on enhanced four plane feature for realistic Chinese handwriting. In: 19th ICPR, 2008Google Scholar
  18. T-H. Su, T-W. Zhang, Z-W. Qiu , D-J. Guan, Gabor-based recognizer for Chinese handwriting from segmentation-free strategy. In: Proceedings of the 12th International Conference on Computer Analysis of Images and Patterns, 2007Google Scholar
  19. A.O. Thomas, A.I. Rusu, V. Govindaraju, Synthetic handwritten CAPTCHAs. Pattern Recogn. 42(12), 3365–3373 (2009)zbMATHCrossRefGoogle Scholar
  20. T. Varga, Off-line cursive handwriting recognition using synthetic training data. PhD thesis, University of Bern, Bern, 2006Google Scholar
  21. Q-F. Wang, F. Yin, C-L. Liu, Integrating language model in handwritten Chinese text recognition. In: International Conference on Document Analysis and Recognition. 2009, pp 1036–1040Google Scholar

Copyright information

© The Author(s) 2013

Authors and Affiliations

  1. 1.Computer ScienceHarbin Institute of TechnologyHarbinP.R. China

Personalised recommendations