Multilevel Classifier in Recognition of Handwritten Arabic Characters | Science Publications

Journal of Computer Science

Enas F. AlRawashdeh¹  Rawan I. Zaghloul¹  Dojanah M.K. Bader¹ 

关键词: Pattern recognition;    binary image;    Arabic alphabet;    Arabic language;    isolated letters;    Arabic script;    OCR system;    radon transform;    Optical Recognition System (OCR);   

DOI  :  10.3844/jcssp.2011.512.518

学科分类：计算机科学（综合）

来源: Science Publications

【 摘 要 】

Problem statement: Arabic offline handwriting recognition is considered one of the mostchallenging topics. This is probably caused by the fact that Arabic recognition system faced manyproblems during the development stage. It faces the usual problems of character recognition in general,in addition to the problems that are specific to Arabic language only. The aim of this study was to builda classifier to solve Arabic text ambiguity; to be used in text recognition applications. Approach: Amultilevel classifier based on pattern recognition techniques, is proposed. The proposed classifierwas implemented using MATLAB and also tested with a large sample of handwritten datasets.Results: Pattern recognition techniques are used to identify Arabic handwritten text. After testing, therecognition rates reached {93, 84, 89 and 85%} for the isolated letters, letters at the beginning, at themiddle and at the end of the word respectively. Conclusion: Even that the Arabic letters change theirshape depending on their position in a word, the proposed classifier, using the powerful set of features,is proved to be effective in the recognition of Arabic letters.

【 授权许可】

Unknown