Title :
Characters´ boundaries based segmentation for online Arabic handwriting
Author :
Mustafa, Mohammed Elhafiz ; Alshafy, Hozeifa Adam Abd
Author_Institution :
Coll. of Comput. Sci. & Inf. Technol., Sudan Univ. of Sci. & Technol., Khartoum, Sudan
Abstract :
This paper presents a segmentation approach for online Arabic handwriting. The approach achieves the segmentation based on the boundaries detection of the handwritten characters which included in the handwriting. The approach relies on an algorithm of characters´ boundaries detection. The algorithm application is preceded by a feature extraction stage which in turn preceded by a preprocessing stage. The obtained results show a significant improvement toward the recognition of online Arabic handwriting.
Keywords :
feature extraction; handwritten character recognition; image segmentation; natural language processing; character boundary based segmentation; feature extraction stage; feature preprocessing stage; handwritten character boundary detection; online Arabic handwriting; Educational institutions; Feature extraction; Handwriting recognition; Hidden Markov models; Smoothing methods; Vectors; Writing; CBD; On-line Arabic handwriting; feature extraction; preprocessing;
Conference_Titel :
Computing, Electrical and Electronics Engineering (ICCEEE), 2013 International Conference on
Conference_Location :
Khartoum
Print_ISBN :
978-1-4673-6231-3
DOI :
10.1109/ICCEEE.2013.6633952
