This Article 
 Bibliographic References 
 Add to: 
A Fast Statistical Mixture Algorithm for On-Line Handwriting Recognition
December 1994 (vol. 16 no. 12)
pp. 1227-1233

The automatic recognition of online handwriting is considered from an information theoretic viewpoint. Emphasis is placed on the recognition of unconstrained handwriting, a general combination of cursively written word fragments and discretely written characters. Existing recognition algorithms, such as elastic matching, are severely challenged by the variability inherent in unconstrained handwriting. This motivates the development of a probabilistic framework suitable for the derivation of a fast statistical mixture algorithm. This algorithm exhibits about the same degree of complexity as elastic matching, while being more flexible and potentially more robust. The approach relies on a novel front-end processor that, unlike conventional character or stroke-based processing, articulates around a small elementary unit of handwriting called a frame. The algorithm is based on (1) producing feature vectors representing each frame in one (or several) feature spaces, (2) Gaussian K-means clustering in these spaces, and (3) mixture modeling, taking into account the contributions of all relevant clusters in each space. The approach is illustrated by a simple task involving an 81-character alphabet. Both writer-dependent and writer-independent recognition results are found to be competitive with their elastic matching counterparts.

[1] C. C. Tappert, C. Y. Suen, and T. Wakahara, "The state of the art in on line hand writing recognition,"IEEE Trans. Patt. Anal. Mach. Intel., vol. 12, no. 8, pp. 787-808, 1990.
[2] H. L. Teulings and L. R. Schomaker, "Unsupervised learning of prototype allographs in cursive-script recognition using invariant handwriting features," inProc. 2nd Int. Workshop on Frontiers in Handwriting Recognition, Bonas, France, September 1991, pp. 45-55.
[3] P. Morasso, J. Kennedy, E. Antonj, S. Di Marco, and M. Dordini, "Self-organization of an allograph lexicon," inProc. Int. Neural Network Conf., Paris, France, 1990. Dordrecht: Kluwer Academic Publ. 1990, pp. 141-144.
[4] L. R. Bahl, F. Jelinek and R. L. Mercer, "A Maximum Likelihood Approach to Continuous Speech Recognition,"IEEE Trans. Patt. Anal. Mach. Intel., Vol. PAMI-5, no. 2, pp. 179-190, Mar. 1983.
[5] C. C. Tappert, "Speed, Accuracy, and Flexibility Trade-Offs in On-Line Character Recognition,"Intl. Journal of Pattern Recognition and Artificial Intelligence, Vol. 5, No. 1-2, pp. 79-95, 1991.
[6] E. J. Bellegarda, J. R. Bellegarda, D. Nahamoo and K. S. Nathan, "A Probabilistic Framework for On-Line Handwirting Recognition," inProc. 3rd Intl. Workshop on Frontiers in Handwriting Recognition, Buffalo, NY, pp. 225-234, May 1993.
[7] T. Fujisaki, K. Nathan, W. Cho and H. Beigi, "On-Line Unconstrained Hand-writing Recognition by a Probabilistic Method," inProc. 3rd Intl. Workshop on Frontiers in Handwriting Recognition, Buffalo, NY, pp. 235-241, May 1993.
[8] J. R. Bellegarda and D. Nahamoo, "Tied Mixture Continuous Parameter Modeling for Speech Recognition," inIEEE Trans. Acoust. Speech, Signal Processing, Vol. 38, No. 12, pp. 2033-2045, Dec. 1990.
[9] J. A. Hartigan,Clustering Algorithms, J. Wiley: Ed., 1975.
[10] L. E. Baum, T. Petrie, G. Soules and N. Weiss, "A maximisation technique occuring in the statistical analysis of probabilistic functions of Markov chains,"Ann. Mathematical Statistics, vol. 41, no. 1, pp. 164-171, 1970.
[11] L. E. Baum, "An inequality and associated maximization technique in statistical estimation for probabilistic functions of Markov processes,"Inequalities, vol. 3, pp. 1-8, 1972.
[12] N. J. Nilsson,Problem Solving Methods in Artificial Intelligence. New York: McGraw-Hill, 1971.
[13] F. Jelinek, "A fast sequential decoding algorithm using a stack," inIBM J. Res. Dev., vol. 13, pp. 675-685, 1969.

Index Terms:
handwriting recognition; statistical analysis; information theory; online operation; fast statistical mixture algorithm; online handwriting recognition; information theory; unconstrained handwriting; cursively written word fragments; discretely written characters; elastic matching; probabilistic framework; complexity; front-end processor; frame representation; feature vectors; feature spaces; Gaussian K-means clustering; mixture modeling; 81-character alphabet; writer-dependent recognition; writer-independent recognition; statistical modeling; frame-based processing; mixture output distributions
E.J. Bellegarda, J.R. Bellegarda, D. Nahamoo, K.S. Nathan, "A Fast Statistical Mixture Algorithm for On-Line Handwriting Recognition," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, no. 12, pp. 1227-1233, Dec. 1994, doi:10.1109/34.387484
Usage of this product signifies your acceptance of the Terms of Use.