Title :
Shape analysis of Pashto script and creation of image database for OCR
Author :
Wahab, Mehreen ; Amin, Hassan ; Ahmed, Farooq
Author_Institution :
Nat. Univ. of Comput. & Emerging Sci., Pakistan
Abstract :
Development of optical character recognition for the cursive script such as Pashto requires detailed knowledge of shape variation within Pashto script. The development of image dataset is essential for training/testing of various OCR approaches. This paper outlines various features of Pashto script, and describes the development of an image dataset for an optical character recognition system.
Keywords :
learning (artificial intelligence); optical character recognition; shape recognition; visual databases; OCR testing; OCR training; Pashto script analysis; image database; optical character recognition; shape analysis; Character recognition; Handwriting recognition; Humans; Image analysis; Image databases; Natural languages; Optical character recognition software; Optical sensors; Shape; Writing;
Conference_Titel :
Emerging Technologies, 2009. ICET 2009. International Conference on
Conference_Location :
Islamabad
Print_ISBN :
978-1-4244-5630-7
Electronic_ISBN :
978-1-4244-5631-4
DOI :
10.1109/ICET.2009.5353160