Offline cursive word recognition using continuous density hidden Markov models trained with PCA or ICA features

Vinciarelli, A. and Bengio, S. (2002) Offline cursive word recognition using continuous density hidden Markov models trained with PCA or ICA features. In: Pattern Recognition, 2002. Proceedings. 16th International Conference on, Quebec, Canada, 11-15 August 2002, pp. 81-84. (doi:10.1109/ICPR.2002.1047800)

Full text not currently available from Enlighten.

Publisher's URL: http://dx.doi.org/10.1109/ICPR.2002.1047800

Abstract

This work presents an offline cursive word recognition system dealing with single writer samples. The system is based on a continuous density hidden Markov model trained using either the raw data, or data transformed using principal component analysis or independent component analysis. Both techniques significantly improved the recognition rate of the system. Preprocessing, normalization and feature extraction are described as well as the training technique adopted. Several experiments were performed using a publicly available database. The accuracy obtained is the highest presented in the literature over the same data.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Vinciarelli, Professor Alessandro
Authors: Vinciarelli, A., and Bengio, S.
College/School:College of Science and Engineering > School of Computing Science

University Staff: Request a correction | Enlighten Editors: Update this record