A neuro-fuzzy inference engine for Farsi numeral characters recognition

Information Systems Dept., DeGroote School of Business, McMaster University, Hamilton, ON, Canada
Expert Systems with Applications (Impact Factor: 1.97). 09/2010; 37(9):6327-6337. DOI: 10.1016/j.eswa.2010.02.088
Source: DBLP

ABSTRACT Character recognition of Farsi and Arabic texts as an open and demanding problem needs to encounter sophisticated specifications of the characters such as their shapes, continuity, dots and also, different fonts. Utilizing fuzzy set theory as a tolerant approach toward uncertainty and vagueness and artificial neural networks as a machine learning method in this paper, we propose a neuro-fuzzy inference engine to recognize the Farsi numeral characters. This engine takes holistic approach of character recognition through the comparison of the unknown character’s features with the features of the existing characters that itself is characterized through Mamdani inference engine on fuzzy rules which is largely enhanced with a multi layer perceptron neural network’s learning on features of the different fonts’ characters which leads to more comprehensive recognition of Farsi numeral characters in the proposed system. Having applied this novel engine on a dataset of unknown numeral characters consisted of 33 different Farsi fonts, it yielded more accurate results than the corresponding researches. The recognition rates of unknown numeral characters are greater than 97% except for Farsi character 4, so as the proposed schema could not score a better result than 95% for this numeral character which implies its recognition is still in need of more enhancement.

1 Follower
  • [Show abstract] [Hide abstract]
    ABSTRACT: The design and implementation of a local area network for the Department of Mathematics and Computer Science at Western Carolina University are presented. This was part of a project to establish one of many local area networks which will eventually be connected by a campus backbone. The various design stages and work details that should be addressed before installing a local area network are discussed. A schematic design of the layout of the network is included
    System Theory, 1993. Proceedings SSST '93., Twenty-Fifth Southeastern Symposium on; 04/1993
  • [Show abstract] [Hide abstract]
    ABSTRACT: The performance of a code-division multiple-access (CDMA) cellular system for different RF modulation techniques, such as quadrature phase-shift keying (QPSK) and differential phase-shift keying (DPSK), is analyzed. The analysis emphasizes how voice quality deteriorates with increasing system load and determines the maximum acceptable load and minimum acceptable E <sub>b</sub>/ N <sub>0</sub> for producing acceptable quality voice
    System Theory, 1993. Proceedings SSST '93., Twenty-Fifth Southeastern Symposium on; 04/1993
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: So far, most of the evidential distance and similarity measures proposed in the Dempster-Shafer theory literature have been based on the basic belief assignment function, so as the belief and plausibility functions as two main results of the theory are not directly used in this regard. In this paper, a new evidential distance measure is proposed based on these functions according to nearest neighborhood concept. After assigning basic belief values to propositions and constructing the belief and plausibility functions or the belief interval, this evidential distance measure compares the similarity between the unknown pattern and class belief intervals. For this purpose, we rst acquire the belief and plausibility functions or the belief intervals and then the distance between the belief intervals of uncertain pattern feature vectors and samples are calculated. We applied this novel distance measure to the bacillus colonies recognition and coronary heart disease patients classiication problems to examine the proposed measure capability in contrast to other evidential measures. Our experiment illustrates that the belief interval distance measure yields the accuracy rates of 91.66 and 92.45 percent for unknown bacillus patterns recognition and coronary heart disease patients classiication, respectively, which in contrast to other evidential measures shows superior performance.
    Scientia Iranica 01/2011; 17. · 0.84 Impact Factor
Show more


Available from