A Study of Discriminative Training for HMM-Based Online Handwritten Chinese/Japanese Character Recognition

We present a study of discriminative training of classifiers using both maximum mutual information (MMI) and minimum classification error (MCE) criteria for online handwritten Chinese/Japanese character recognition based on continuous-density hidden Markov models. It is observed that MCE-trained classifiers can achieve a much higher recognition accuracy than that of MMI-trained ones. Benchmark results of MCE-trained classifiers for simplified Chinese, traditional Chinese and Japanese characters are reported on three recognition tasks with a vocabulary of 9119, 20924, and 12333 characters respectively.

[1]  Qiang Huo,et al.  A Study of Feature Design for Online Handwritten Chinese Character Recognition Based on Continuous-Density Hidden Markov Models , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[2]  S. Katagiri,et al.  Discriminative Learning for Minimum Error Classification , 2009 .

[3]  Brian Kan-Wing Mak,et al.  Subspace distribution clustering hidden Markov model , 2001, IEEE Trans. Speech Audio Process..

[4]  Yongqiang Wang,et al.  Sample-separation-margin based minimum classification error training of pattern classifiers with quadratic discriminant functions , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[5]  Biing-Hwang Juang,et al.  Minimum classification error rate methods for speech recognition , 1997, IEEE Trans. Speech Audio Process..

[6]  Zhen-Long Bai,et al.  A study on the use of 8-directional features for online handwritten Chinese character recognition , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[7]  Scott E. Fahlman,et al.  An empirical study of learning speed in back-propagation networks , 1988 .

[8]  Biing-Hwang Juang,et al.  Discriminative learning for minimum error classification [pattern recognition] , 1992, IEEE Trans. Signal Process..

[9]  Allen Gersho,et al.  Vector quantization and signal compression , 1991, The Kluwer international series in engineering and computer science.

[10]  Jonathan Le Roux,et al.  Discriminative Training for Large-Vocabulary Speech Recognition Using Minimum Classification Error , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[11]  Yongqiang Wang,et al.  Design Compact Recognizers of Handwritten Chinese Characters Using Precision Constrained Gaussian Models, Minimum Classification Error Training and Parameter Compression , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[12]  Hiroshi Sako,et al.  Discriminative learning quadratic discriminant function for handwriting recognition , 2004, IEEE Transactions on Neural Networks.

[13]  Steve Young,et al.  The HTK book , 1995 .

[14]  Yong Ge,et al.  A study on the use of CDHMM for large vocabulary off-line recognition of handwritten Chinese characters , 2002, Proceedings Eighth International Workshop on Frontiers in Handwriting Recognition.

[15]  Fumitaka Kimura,et al.  Modified Quadratic Discriminant Functions and the Application to Chinese Character Recognition , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  K. Ishigaki.  A Top-Down On-line Handwritten Character Recognition Method via the Denotation of Variation , 1988 .

[17]  Yongqiang Wang,et al.  A Study of Designing Compact Recognizers of Handwritten Chinese Characters Using Multiple-Prototype Based Classifiers , 2010, 2010 20th International Conference on Pattern Recognition.

[18]  Tingting. He,et al.  A study on several problems in online handwritten Chinese character recognition , 2008 .