You are here

Optical Character Recognition for Arabic Ruq

Award Information
Agency: Department of Defense
Branch: Army
Contract: W15P7T-08-C-G201
Agency Tracking Number: A072-095-1078
Amount: $69,853.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: A07-095
Solicitation Number: 2007.2
Timeline
Solicitation Year: 2007
Award Year: 2007
Award Start Date (Proposal Award Date): 2007-12-20
Award End Date (Contract End Date): 2008-06-20
Small Business Information
727 Airport Boulevard
Ann Arbor, MI 48108
United States
DUNS: 197187602
HUBZone Owned: No
Woman Owned: Yes
Socially and Economically Disadvantaged: Yes
Principal Investigator
 Glenn Beach
 Research Engineer
 (734) 668-2567
 proposals@cybernet.com
Business Contact
 Charles Jacobus
Title: President
Phone: (734) 668-2567
Email: proposals@cybernet.com
Research Institution
N/A
Abstract

In the ongoing efforts in Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF), as well as the continuing Global War on Terror (GWOT), U.S. forces gain valuable information from captured documents. While these documents are typically written in Arabic, there are not enough U.S. personnel trained in reading Arabic to expediently read and understand the captured information. Since the information has the most value while it is still fresh, the U.S. military has started to use Arabic based optical character recognition (OCR) to quickly convert the Arabic text into English. While these commercial OCR packages work well with machine-produced documents, their performance is quickly and significantly degraded on more typical hand-written Arabic documents. The problems are compounded when the original document contains even small levels of noise or other sources of image degradation. In order to more quickly and effectively process collected information there is a need for an automated system to process this handwritten text. We propose to leverage our experience with developing previous image processing and handwriting recognition systems to develop a system for performing optical character recognition (OCR) on the most commonly used Arabic script, Ruq’ah

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government