Aboaisha, Hosain, Xu, Zhijie and El-Feghi, Idris (2012) An investigation on efficient feature extraction approaches for Arabic letter recognition. In: Proceedings of The Queen’s Diamond Jubilee Computing and Engineering Annual Researchers’ Conference 2012: CEARC’12. University of Huddersfield, Huddersfield, pp. 80-85. ISBN 978-1-86218-106-9

Invariant features play an essential role in many pattern recognition applications due to their robustness to different environmental settings. In this research, two of the invariant moments, the Zernike Moment (ZM) and the Legendre Moment (LM), have been investigated for their suitability and computational efficiency in Arabic
letter recognition. This paper starts with an introduction on Arabic letter characteristics before moving on to a literature review of current letter recognition strategies and systems. The concepts and algorithms of the ZM and
LM techniques are then examined. To validate the new approach, a prototype system has been developed with several experiments carried out using a standard database called IF/ENIT which contains handwritten Tunisian town names and this database is used by many research groups working on recognition systems.

Cover page
Cover_pages.pdf - Published Version

Download (1MB) | Preview
Additional paper
H_Aboaisha_Paper.pdf - Published Version

Download (193kB) | Preview


Downloads per month over past year

Downloads per month over past year for

Downloads per month over past year for

Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email