Abstract
Problem statement: Significant movement has been made in handwriting recognition technology over the last few years. Up until now, Arabic handwriting recognition systems have been limited to small and medium vocabulary applications, since most of them often rely on a database during the recognition process. The facility of dealing with large database, however, opens up many more applications. Approach: This study presented a complete system to recognize off-line Arabic handwriting image and Arabic handwriting and printed text database AHPD-UTM that used to implement and test the system. That system start from preprocessing and segmentation phases that deepened on thinning the image and found the V and H projection profile until recognition phase by genetic algorithm. Results: The genetic algorithm stand on feature extraction algorithm that defined six feature for each segment beak. The system can be recognized Arabic handwriting with 87% accuracy. The confusion and rejection rates are 8.4, those causes for several problems like characters with broken loops and character segmentation problem. Conclusion: Peak connection solved some of the segmentation problems and helped to provide better accuracy.
Highlights
Character Recognition (CR) mechanization occupies an intensive research region of the pattern recognition research area
This study describes an extended version of an offline handwritten Arabic word recognition system based on Feature extraction approach and Genetic Algorithms
Experimental and result: In order to check the accuracy of the Arabic handwriting recognition system using genetic algorithm, handwriting samples in AHPD-UTM has been used
Summary
Character Recognition (CR) mechanization occupies an intensive research region of the pattern recognition research area. CR automation means translating images of characters into an editable text, in other words, it represents an attempt to simulate the human reading process. In other said handwriting recognition is a very challenging task due to the existence of many difficulties such as the high variability of the handwritten styles and shapes, uncertainty of human writing, writing skew or slant, segmentation of the words into characters and the size of the lexicon (Amin and Kavianifar, 1997). Only the image of the handwriting is available, while in the on-line case temporal information such as pen tip coordinates, as a function of time, is available. Many applications require offline HWR capabilities such as bank processing, mail sorting, document archiving, commercial form-reading and office automation. Off-line HWR remains an open problem, in spite of a dramatic boost of research (Koerich et al, 2003; Plamondon and Srihari, 2000; Vinciarelli, 2002) in this field and the latest improvement in recognition methodologies (El-Yacoubi et al, 1999; Vinciarelli et al, 2004; Lorigo and Govindaraju, 2006)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
