The reduction of memory and the improvement of recognition rate for HMM on-line handwriting recognition

The purpose of this project is two fold. The first purpose is to reduce the memory size of our previous handwriting recognition algorithm based on an HMM using self-organizing map (SOM) density tying. The second is to improve recognition capability by incorporating additional information. SOM density tying reduced the dictionary size to 1/7 of the original size, with a recognition rate of 90.45%, only slightly less than the original recognition rate of 91.51%. Our additional feature increased recognition capability to 91.34%.

[1]  H. Shu.  On-Line Handwriting Recognition Using Hidden Markov Models , 2002 .

[2]  Shigeki Sagayama,et al.  Substroke approach to HMM-based on-line Kanji handwriting recognition , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[3]  JinHyung Kim,et al.  Data-driven Design of HMM Topology for On-line Handwriting Recognition , 2000 .

[4]  Christian Viard-Gaudin,et al.  From Off-line to On-line Handwriting Recognition , 2004 .

[5]  K. Takahashi,et al.  A fast HMM algorithm for on-line handwritten character recognition , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[6]  Robert Sabourin,et al.  Fast two-level Viterbi search algorithm for unconstrained handwriting recognition , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[7]  Sargur N. Srihari,et al.  On-Line and Off-Line Handwriting Recognition: A Comprehensive Survey , 2000, IEEE Trans. Pattern Anal. Mach. Intell..