Open Source Arabic OCR projects

The Siragi project  http://www.arabeyes.org/project.php?proj=Siragi started in 2005. It is part of the Arabic Unix open source project.

Commercial Arabic OCR

There are a number of commercial Arabic OCR engines:

• Sakhr's Automatic Reader (Amazon link: http://www.amazon.com/Automatic-Reader-Multilingual-OCR-Gold/dp/B0002A5D5U). Pretty expensive: ~1500$ for Shrink wrap and ~5000$ for the SDK.

• Readiris from IRIS http://www.irislink.com/c2-561/Readiris-Pro-11-Middle-East---Arabic-OCR-Software.aspx ~500$

• Verus from NovoDynamics http://www.novodynamics.com/verus_stand.htm ~1300$

• Caere announded an Arabic version of its well known Omnipage engine in 1996 http://www.encyclopedia.com/doc/1G1-18082394.html, but it does not seem to be publicly available now.

Public Data Sets 

• DARPA/SAIC database: "R. Davidson and R. Hopely. Arabic and Persian OCR training and test data sets. In Proc. of Symp. on Document Image Understanding Technology, Annapolis, MD, May 1997".

• ERIM Arabic Database 

• IFN/ENIT  

Related Software

• ArabEyes

• Siragi 

Related Publications

• http://portal.acm.org/citation.cfm?id=876867.877654
• http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel5/7569/20622/00953967.pdf
• http://ieeexplore.ieee.org/Xplore/login.jsp?url=/iel5/9457/30014/01374482.pdf
• http://tev.itc.it/OCR/Products.html
• http://citeseer.ist.psu.edu/26451.html
• http://www.smi.uib.no/ksv/ArabOCR.html
• An Omnifont Open-Vocabulary OCR System for English and Arabic
• ICDAR 2005 Arabic Handwriting Recognition Competition
• http://hal.inria.fr/action/open_file.php?url=http://hal.inria.fr/docs/00/11/27/54/PDF/cr112431769559.pdf&docid=112754