INCORPORATING A NEW RELATIONAL FEATURE IN ARABIC ONLINE HANDWRITTEN CHARACTER RECOGNITION

Sara Izadi, Ching Y. Suen

Abstract

Artificial neural networks have shown good performance in classification tasks. However, models used for learning in pattern classification are challenged when the differences between the patterns of the training set are small. Therefore, the choice of effective features is mandatory for obtaining good performance. Statistical and geometrical features alone are not suitable for recognition of hand printed characters due to variations in writing styles that may result in deformations of character shapes. We address this problem by using a relational context feature combined with a local descriptor for training a neural network-based recognition system in a user-independent online character recognition application. Our feature extraction approach provides a rich representation of the global shape characteristics, in a considerably compact form. This new relational feature provides a higher distinctiveness and increases robustness with respect to character deformations. While enhancing the recognition accuracy, the feature extraction is computationally simple. We show that the ability to discriminate in Arabic handwriting characters is increased by adopting this mechanism in feed forward neural network architecture. Our experiments on Arabic character recognition show comparable results with the state-of-the-art methods for online recognition of these characters.

References

  1. Bahlmann, C., Haasdonk, B., and Burkhardt, H. (2002). On-line handwriting recognition with support vector machines - a kernel approach.
  2. Belongie, S., Malik, J., and Puzicha, J. (2002). Shape matching and object recognition using shape contexts. IEEE Trans. Pattern Anal. Mach. Intell., 24(4):509- 522.
  3. Mezghani, N., Cheriet, M., and Mitiche, A. (2003). Combination of pruned kohonen maps for on-line aracic characters recognition.
  4. Mezghani, N., Mitiche, A., and Cheriet, M. (2005). A new representation of shape and its use for high performance in online arabic character recognition by an associative memory. IJDAR, 7(4):201-210.
  5. Oh, J. and Geiger, D. (2000). An on-line handwriting recognition system using fisher segmental matching and hypotheses propagation network. IEEE Conference on Computer Vision and Pattern Recognition, 2:343 - 348.
  6. Verma, B., Lu, J., Ghosh, M., and Ghosh, R. (2004). A feature extraction technique for online handwriting recognition. IEEE International Joint Conference on Neural Networks, 2:1337 - 1341.
Download


Paper Citation


in Harvard Style

Izadi S. and Y. Suen C. (2008). INCORPORATING A NEW RELATIONAL FEATURE IN ARABIC ONLINE HANDWRITTEN CHARACTER RECOGNITION . In Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008) ISBN 978-989-8111-21-0, pages 559-562. DOI: 10.5220/0001087005590562


in Bibtex Style

@conference{visapp08,
author={Sara Izadi and Ching Y. Suen},
title={INCORPORATING A NEW RELATIONAL FEATURE IN ARABIC ONLINE HANDWRITTEN CHARACTER RECOGNITION},
booktitle={Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008)},
year={2008},
pages={559-562},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001087005590562},
isbn={978-989-8111-21-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008)
TI - INCORPORATING A NEW RELATIONAL FEATURE IN ARABIC ONLINE HANDWRITTEN CHARACTER RECOGNITION
SN - 978-989-8111-21-0
AU - Izadi S.
AU - Y. Suen C.
PY - 2008
SP - 559
EP - 562
DO - 10.5220/0001087005590562