Title :
Global features for offline recognition of handwritten Arabic literal amounts
Author :
El-Melegy, Moumen T. ; Abdelbaset, Asmaa A.
Author_Institution :
Assiut Univ., Assiut
Abstract :
The domain of automatic handwriting recognition has a variety of applications in real world problems such as cheque processing, office automation and data entry applications. In this paper an approach for describing the role of holistic structural features in the recognition of offline handwritten Arabic literal amounts has been presented. Our proposed system attempts to recognize words from their overall shape. The extracted features are presented to several classifiers. The system is trained and tested using an Arabic handwritten database.
Keywords :
feature extraction; handwriting recognition; image classification; Arabic handwritten database; automatic handwriting recognition; cheque processing; data entry applications; handwritten Arabic literal amounts; holistic structural features; office automation; offline recognition; Character recognition; Classification tree analysis; Data mining; Handwriting recognition; Hidden Markov models; Humans; Neural networks; Office automation; Shape; Writing; Arabic Word Recognition; Global Features; Handwritten Recognition; Literal Amount Recognition;
Conference_Titel :
Information and Communications Technology, 2007. ICICT 2007. ITI 5th International Conference on
Conference_Location :
Cairo
Print_ISBN :
978-1-4244-1430-7
DOI :
10.1109/ITICT.2007.4475631
