Handwritten similar Chinese characters recognition based on multi-class pair-wise support vector machines

Author

Feng, Jun ; Chen, Dong-E

Author_Institution

Dept. of Comput. Sci., Shijiazhuang Railway Inst., China

Volume

7

fYear

2005

fDate

18-21 Aug. 2005

Firstpage

4405

Abstract

The existence of lots of similar characters is a key factor affecting classification performance of off-line handwritten Chinese characters recognition. This paper studies on handwritten similar Chinese characters recognition problem based on support vector machines. The features of normalized Chinese character are extracted by the method of wavelet transform and elastic meshing. The classification performances of similar Chinese characters are compared based on max voting, fuzzy pair-wise SVM and directed acyclic graph.

Keywords

directed graphs; feature extraction; fuzzy set theory; handwritten character recognition; natural languages; pattern classification; support vector machines; wavelet transforms; directed acyclic graph; elastic meshing; feature extraction; fuzzy pair-wise SVM; handwritten similar Chinese character recognition; max voting; multiclass pair-wise support vector machines; pattern classification; wavelet transform; Character recognition; Feature extraction; Handwriting recognition; Machine learning; Neural networks; Statistics; Support vector machine classification; Support vector machines; Voting; Wavelet transforms; Handwritten similar Chinese characters; multi-class classifier; pair-wise based combination strategy; support vector machines (SVM); wavelet transform and elastic meshing;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on

Conference_Location

Guangzhou, China

Print_ISBN

0-7803-9091-1

Type

conf

DOI

10.1109/ICMLC.2005.1527714

Filename

1527714