A New Radial Basis Function Artificial Neural Network based Recognition for Kurdish Manuscript

A New Radial Basis Function Artificial Neural Network based Recognition for Kurdish Manuscript

Behnam Zebardast, Isa Maleki
Copyright: © 2013 |Pages: 16
DOI: 10.4018/ijaec.2013100105
(Individual Articles)
No Current Special Offers


During recent decades, recognizing letters was a considerable discussion for artificial intelligence researchers and recognize letters due to the variety of languages and different approaches have many challenges. The Artificial Neural Networks (ANNs) are framed based on particular application such as recognition pattern and data classification through learning process is configured. So, it is a proper approach to recognize letters. Kurdish language has two popular handwritings based on Arabic and Latin. In this paper, Radial Basis Function (RBF) of ANNs is used to recognize Kurdish-Latin manuscripts. Although, the authors' proposed method is also used to recognize the letters of all Latin languages which include English, Turkish and etc. are used. The authors implement RBF of ANNs in MATLAB environment. In this paper, the efficiency criteria is supposed to minimize the Mean Square Error (MSE) to recognize Kurdish letters and maximize recognition accuracy of Kurdish letters in training and testing stage of RBF of ANNs. The recognition accuracy in training and testing stages are 100% and 96.7742%, respectively.
Article Preview

1. Introduction

Nowadays, machines have many applications in different contexts with the ability to recognize pattern. The examples of these machines are those which capable of reading alphabetical symbols. In fact, these provide economical alternatives to user directory method. These programs save time and cost and cause less errors. One of these machines which capable of learning is ANNs (Demuth & Beale, 2002). The ANNs play an important role in pattern recognition applications. The recognition systems are used in many contexts which had different natures. One of these contexts is letters recognition. During last decades, considerable developments are achieved in this feature such as automatic reading of postal address, bank checks and so on. Letters recognition is one of the issues which don't have any particular rule or algorithm and it can be used different methods to design its systems and each one has its own special advantages and disadvantages. Letters recognition is a process which must be performed several repetitions in it. This repetition process must be occurred to get desirable results (Rashnodi, Sajedi, & Abadeh, 2011; Meisels, Kandel, & Gecht, 1989).

Letters recognition is one of the fields of pattern recognition, which has been the subject of considerable research (Ahangar & Ahangar, 2009). Many reports are published about word recognition in different languages such as Chinese (Yeung, 1994), Japanese (Yamada, Kami, Temma, & Tsukumo, 1989), English (Morasso, 1989; Smith, Baurgoin, Sims, & Voorhees, 1994), Arabic (Almualim & Yamaguchi, 1987; Abuhaiba, Mahmoud, & Green, 1994; Gharehchopogh & Ahmadzadeh, 2012) and Farsi (Ahangar & Ahangar, 2009; Gharehchopogh & Ahmadzadeh, 2012). But yet, recognizing Kurdish letters due to the some occurred issues are in recognizing these letters aren't performed. It is a very complicated problem which includes many changes in writing styles, size and different sides of different languages' characters. Because, writing characters as matrix in different sizes and make them into the particular form is a difficult task. These changes can be repeated infinitely (Gharehchopogh & Ahmadzadeh, 2012; Gharehchopogh, 2011) and ANNs had proper efficiency to solve this problem. The ANNs is adopted as a new technology in computer science which supposed to simulate human brain neurons behavior. It is mostly used in many cases in the recent years such as ANNs capabilities in recognition, estimation and prediction (Gharehchopogh, 2011).

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing