Classification of printed Chinese characters by using neural network

  • Authors:

  • Attaullah Khawaja;Abdul Fattah Chandio;Alataf Rajpar;Ali Raza Jafri

  • Affiliations:

  • Department of Electronics Engineering, Beijing Institute of Technology, Beijing, P.R.China;Department of Computer Sciences, Beijing Institute of Technology, Beijing, P.R. China;Department of Electro, Mechanical Engineering, Beijing Institute of Technology, Beijing, P.R. China;Department of Industrial and Manufacturing, NED University of Engineering and Technology, Karachi, Pakistan

  • Venue:
  • SIP'06 Proceedings of the 5th WSEAS international conference on Signal processing
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

Unlike human brains that can identify and memorize the characters like letters or digits, computers treat them as binary graphics. Therefore, algorithms are necessary to identify and recognize each character. This article presents a technique to recognize Printed Chinese characters by using the Probabilistic Neural Networks (PNN). The article also discusses Chinese character recognition using projection feature extraction method. In this paper we present the design of neural network after extraction of features that is able to recognize and classify samples of Chinese characters. Our first target is to train the proper neural network to recognize them, the second and the most challenging target is to train this neural network to recognize these characters after they are corrupted by different types Noise. We consider three kinds of noise, Gaussian, speckle (uniformly distributed random noise) and Impulse noise, since several applications suffer. Results show that this method has the advantages of fast processing speed, accurate recognition rate and strongly resisting noise.