Journal of South China University of Technology (Natural Science Edition) ›› 2010, Vol. 38 ›› Issue (6): 78-83.doi: 10.3969/j.issn.1000-565X.2010.06.015

• Computer Science & Technology • Previous Articles     Next Articles

Character Representation of Handwritten Arabic Numerals Based on Wavelet Analysis and EMD

Li He-long1 Wang Wen-bo2  Zhang Guan-xiang1   

  1. 1.Department of Electronic Commerce,South China University of Technology,Guangzhou 510006,Guangdong,China;2.Department of Information and Computing Science,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China
  • Received:2009-07-14 Revised:2010-01-26 Online:2010-06-25 Published:2010-06-25
  • Contact: 李合龙（1977-），男，副教授，博士，主要从事人工智能与电子商务研究． E-mail:hlongli@scut．edu．cn
  • About author:李合龙（1977-），男，副教授，博士，主要从事人工智能与电子商务研究．
  • Supported by:

    国家自然科学基金资助项目（10826053 60825306）; 国家自然科学基金重点项目（U0735004）; 广东省自然科学基金重点项目（07118074）; 华南理工大学中央高校基本科研业务费专项资金资助项目（2009ZM0081 2009ZZ0071 2009ZM0198）

PDF

1979

Abstract:

As the empirical mode decompositon（EMD） can accurately recognize the structure of the original signal,this paper proposes a new feature extraction algorithm of handwritten Arabic numerals based on wavelet transform and EMD.In this algorithm,first,smooth contours of numeral image are obtained by preprocessing the maximum module of the G wavelet transform.Then,an EMD analysis is performed to decompose the normalized curvature sequences into their components,which produces more compact curvature features.Finally,the obtained curvature features are clustered and recognized.Experimental results show that the proposed algorithm is superior to the classic feature extraction algorithm in terms of clustering effect and classifier design ability.

Key words: feature extraction, wavelet transform, empirical mode decomposition（EMD）, curvature