Abstract
This work presents a neural post-processor introducing lexical knowledge in a neural predictive system for on-line word recognition [4]. Each word is modeled by the natural concatenation of letter-models corresponding to the letters composing it. Successive parts of a word trajectory are this way modeled by different Neural Networks. A dynamical segmentation allows to adjust letter-models to the great variability of handwriting encountered in the words. Our system combines Multilayer Neural Networks and Dynamic Programming with an underlying Left-Right Hidden Markov Model (HMM). Training was performed on 7000 words from 9 writers, leading to already good results in the letter-labelling process. These results are significantly improved, at the word level, thanks to the use of the postprocessor.
Preview
Unable to display preview. Download preview PDF.
References
Bengio Y., Le Cun Y., Nohl C., Burges C.: “LeRec: a NN/HMM hybrid for on-line handwriting recognition”, Neural Computation 7, 1289–1303, 1995.
Bercu S., Lorette G.: “On-line handwritten word recognition: an approach based on Hidden Markov Models”, IWFHR 93, pp. 385–390.
Farag R., “Word level recognition of cursive script”, IEEE Trans. on Comp., Vol. C28, pp. 172–175, 1979.
Garcia-Salicetti S., Gallinari P., Dorizzi B., Wimmer Z., Gentric S.: “From characters to words: dynamical segmentation and predictive neural networks”, ICASSP 96, Atlanta, May 1996.
Manke S., Finke M., Waibel A.: “NPen++: a writer-independent, large vocabulary on-line handwriting recognition system”, Proceedings of ICDAR 95, pp. 403–408, 1995.
Juang B-H., Rabiner L.R., “The Segmental K-Means algorithm for estimating parameters of Hidden Markov Models”, IEEE Transactions on Acoustics, Speech, and Signal Processing, Vol. 38, N∘9, pp. 1639–1641, 1990.
Rabiner L.R., Juang B-H.: “Fundamentals of speech recognition”, Prentice Hall Signal Processing Series, 1993.
Schenkel M., Guyon I., Henderson D., “On-line cursive script recognition using Time Delay Neural Networks and Hidden Markov Models”, ICASSP 94, pp. II637–640, Adelaide, 1994.
Schomaker L.R.B., Teulings H.L.: “Stroke versus character-based recognition of on-line connected cursive script”, From Pixels to Features III: Frontiers in Handwriting Recognition, S. Impedovo and J.C. Simon (eds), 1992, Elsevier Science Publishers B.V.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Garcia-Salicetti, S. (1996). A neural lexical post-processor for improved neural predictive word recognition. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_100
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_100
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive