Abstract
In a recent publication [1], we have introduced a neural predictive system for on-line word recognition. Our approach implements a Hidden Markov Model (HMM)-based cooperation of several predictive neural networks. The task of the HMM is to guide the training procedure of neural networks on successive parts of a word. Each word is modeled by the concatenation of letter-models corresponding to the letters composing it. Successive parts of a word 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). In this paper, we present an extension of this model to off-line word recognition. We use on-line data in these off-line experiments, generating a binary image from trajectory data. The feature extraction module then turns each binary image into a sequence of feature vectors, called ‘frames’, combining low-level and high-level features in a new feature extraction paradigm. Some results for word recognition are presented.
Preview
Unable to display preview. Download preview PDF.
7 References
S.Garcia-Salicetti, B.Dorizzi, P.Gallinari, Z.Wimmer, “Adaptive Discrimination in an HMM-Based Neural Predictive System for On-line Word Recognition”, Proceedings of ICPR'96, pp.515–519, Wien, 1996.
S. Garcia-Salicetti, B. Dorizzi, P. Gallinari, A. Mellouk, D. Fanchon, “A Hidden Markov Model extension of a neural predictive system for on-line character recognition”, Proceedings of ICDAR 95, 1995, pp. 50–53.
S. Garcia-Salicetti, P. Gallinari, B. Dorizzi, Z. Wimmer, S. Gentric, “From Characters To Words: Dynamical Segmentation And Predictive Neural Networks”, Proceedings Of Icassp 96, 1996, Atlanta.
R.Seiler, M.Schenkel, F.Eggimann, “Off-line Cursive Handwriting Recognition Compared with On-line Recognition”, Proceedings of ICPR'96, pp.505–509, Wien, 1996.
M.Schenkel, I.Guyon, D.Henderson, “On-line cursive script recognition using Time Delay Neural Networks and Hidden Markov Models”, Proceedings of ICASSP'94, pp. 11637–640, Adelaide, 1994.
Andrew William Senior, “Off-line cursive handwritting recognition using recurrent neural networks”, Thesis of University of Cambridge, 1994.
R.G. Casey, “Moment Normalization of Handprinted Caracters», IBM J. Res. Develop., 1970, p.: 548–557.
M.K.Hu, “Visual invariant pattern recognition by moment invariants”, IRE Transactions on Information Theory, vol.8, pp. 179–187, 1962.
M.Gilloux, M.Leroux, J-M. Bertille, “Strategies for Handwritten Words Recognition Using Hidden Markov Models», Proceedings of ICDAR 93, 1993,pp. 299–304.
Kjersti Aas, Line Eikvil, “Text Page Recognition using Grey-level features and Hidden Markov Models», Pattern Recognition, vol.29. No.6, pp.977–985, 1996
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wimmer, Z., Garcia-Salicetti, S., Dorizzi, B., Gallinari, P. (1997). Off-line cursive word recognition with a hybrid neural-HMM system. In: Murshed, N.A., Bortolozzi, F. (eds) Advances in Document Image Analysis. BSDIA 1997. Lecture Notes in Computer Science, vol 1339. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63791-5_19
Download citation
DOI: https://doi.org/10.1007/3-540-63791-5_19
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63791-2
Online ISBN: 978-3-540-69646-9
eBook Packages: Springer Book Archive