Abstract
Several methods of combination of Multilayer Perceptrons (MLPs) for handwritten character recognition are presented and discussed. Recognition tests have shown that cooperation of neural networks using different features vectors can reduce significantly the overall misclassification error rate. Additionally, the MLPs that are combined are the results of the experiments that were previously performed in order to optimize the recognition process when using a single MLP. So, all the combination methods that are proposed are very easy to carry out. The final recognition system consists of a cascade association of small MLPs, which allows minimization of the overall recognition time while retaining a high recognition rate. This system appears to be 2.5 times faster than the best of the individual MLPs, while offering a recognition rate of 99.8% on unconstrained digits extracted from the NIST 3 database.
Similar content being viewed by others
References
M.D.Garris and R.A.Wilkinson. “Handwritten segmented characters database”, Technical Report Special Database 3, HWSC, National Institute of Standards and Technology, Gaithersburg, MD, USA, 1992.
S. Impedevo, L. Ottaviano and S. Occhingro, “Optical character recognition — a survey”, Int. Journal of Pattern Recognition and Artificial Intelligence, 1991.
A. d'Acierno, C. De Stefano and M. Vento, “A structural character recognition method using neural network”, Proc. First Int. Conf. Document Analysis and Recognition, pp. 803–811, Saint-Malo, France, Sept. 30–Oct. 2, 1991.
A.Filipski and R.Flandrena, “Automated conversion of engineering drawings to CAD form”, Proc. IEEE, Vol. 80(7), pp. 1195–1209, 1992.
C.Y.Suen, C.Nadal, R.Legault, T.A.Mai and L.Lam, “Computer recognition of unconstrained handwritten numerals”, Proc. IEEE, Vol. 80(7), pp. 1162–1181, 1992.
H.Susuki, L.O.Chua and T.Matsumoto, “A CNN handwritten character recogniser”, Int. Journal of Circuit Theory and Applications, Vol. 20(5), 601–612, 1992.
B. Gosselin, “A comparison of different feature extractors for handwritten characters recognition”, in J.P. Veen and M.J. de Ket (eds) Proc. of the ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing, Papendael, The Netherlands, March 1994, STW-Technology Foundation: Utrecht, The Netherlands.
K.Fukunaga, Introduction to Statistical Pattern Recognition, Academic Press: London, 1990.
H.Bourlard and N.Morgan, “Connectionist speech recognition — a hybrid approach”, Kluwer Academic Publishers: Dordrecht, The Netherlands, 1991.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Gosselin, B. Multilayer perceptrons combination applied to handwritten character recognition. Neural Process Lett 3, 3–10 (1996). https://doi.org/10.1007/BF00417783
Issue Date:
DOI: https://doi.org/10.1007/BF00417783