Liu Hong-Shen
College of Computer, Anhui University of Technology, Anhui, 243032, Maanshan
Wang Nan
College of Computer, Anhui University of Technology, Anhui, 243032, Maanshan
Li Hong-Feng
College of Computer, Anhui University of Technology, Anhui, 243032, Maanshan
ABSTRACT
The feature combination is the study object of this study. The concept of feature combination and the problem in it are introduced. In order to study the relationship between features, some features are selected and a research platform of character recognition based on BP network is designed. The recognition rates of these features are obtained on the platform. The correlations of the feathers are researched based on the data of recognition rates and a parameter of describing the correlation among features is proposed. The experimental data shows that the parameters used to describe the correlation conform to the facts.
PDF References Citation
How to cite this article
Liu Hong-Shen, Wang Nan and Li Hong-Feng, 2013. Research on the Correlation among the Features of Character Recognition Based on BP Network. Information Technology Journal, 12: 2901-2904.
DOI: 10.3923/itj.2013.2901.2904
URL: https://scialert.net/abstract/?doi=itj.2013.2901.2904
DOI: 10.3923/itj.2013.2901.2904
URL: https://scialert.net/abstract/?doi=itj.2013.2901.2904
REFERENCES
- Wang, F., X. Zhang and B. Cao, 2007. Nonlinear feature fusion based on kernel fisher discriminant analysis for machine condition monitoring. Proceedings of the IEEE International Conference on Digital Object Identifier, August 18-21, 2007, Jinan, pp: 225-230.
CrossRef - Wu, R., Z. Wei, F. Yin and X. Tang, 2005. Application of feature fusion and similarity classifier in English Recoghition. Comput. Eng. Appl., 16: 55-57.
Direct Link - Xue, Y.J. and Y. Ding, 2010. Vehicle license plate character recognition based on relative entropy function criterion. J. Comput. Appl., 30: 977-979.
CrossRef - Caimao, Y., D. Sun, D. Liu, S.Y. Cho and Y. Zhang, 2010. A research on feature selection and fusion in palmprint recognition. Proceedings of the International Workshop on Emerging Techniques and Challenges for Hand-Based Biometrics, August 22-22, 2010, Istanbul, pp: 1-6.
CrossRef