Prime Recognition Announces PrimeOCR Version 4.0

Redmond, WA - Prime Recognition, a leader in production OCR software, announces a major release of its production OCR software PrimeOCR version 4.0. The release improves PrimeOCR's industry leading performance in OCR accuracy and Adobe® PDF output, and adds support for Asian languages.

New features in PrimeOCR version 4.0 include:

· Improved Recognition Accuracy - The base configuration of PrimeOCR (Level 3) improves character accuracy by 15%. Level 6 accuracy improved by 35%. PrimeOCR also introduces a new Level 7 which increases accuracy by 44% compared with the previous version Level 6. For equivalent accuracy compared to the last release, the processing time of PrimeOCR has decreased by 6-10%.

· Accessible PDF Output - A market first, PrimeOCR generates accessible PDF output from scanned pages that complies with Section 508 standards of the U.S. Rehabilitation Act. PrimeOCR includes tags in accessible PDF output that provides alternate text for image zones, identifies correct text reading order and paragraphs and provides for proper identification of table rows and table columns along with the ability to display reflowed text for image plus hidden text documents. Compliance with Section 508 is becoming an important aspect of searchable PDF output for Federal Government agencies and businesses.

· Support for Asian Languages - PrimeOCR 4.0 includes the ability to recognize Japanese, Korean, Simplified Chinese and Traditional Chinese. OCR results can be exported to PDF or Unicode.

Review this release on-line at http://primerecognition.com/augprime/pressv40.htm.

Improved OCR Accuracy through "Voting'� Technology

Developed for the production OCR marketplace, PrimeOCR is an "omnifont'� Windowsâ„¢ NT/2000/XP based software OCR engine and Application Programming Interface (API) toolkit that uses a combination of proprietary "Voting'� algorithms and technology licensed from the best retail OCR vendors.

PrimeOCR is able to reduce OCR errors by an average of 65-82% over the best conventional OCR products while providing much cleaner data by implementing "Voting" technology. Since manual error correction can account for over 50% of an imaging system cost, by lowering error rates, PrimeOCR can reduce annual operating costs by 35%.

Designed around a fault tolerant architecture, PrimeOCR provides the production OCR marketplace with a reliable OCR processing solution that generates cleaner results than conventional OCR systems.
PrimeOCR has won a number of awards, including two "Product of the Year'� awards from Imaging Magazine, and a "Top 10 of AIIM'�.

Prime Recognition Corporation

Prime Recognition designs and develops advanced machine print character recognition software for the production OCR marketplace. Prime Recognition's goal is to facilitate the acceptance of OCR by reducing errors through the use of sophisticated "Voting'� technology.
Prime Recognition worldwide customers include Service Bureaus, System Integrators, University/Library systems, companies in the Banking, Insurance, Legal, Manufacturing, Medical, and Transportation industries. However, the largest customer segment is Government, Prime's largest customer is the U.S. Patent & Trademark office, with a number of installations in Defense Department applications, and other government agencies.

Prime Recognition was founded in 1994 and is headquartered in Woodinville, Washington (on Seattle's "Eastside'�, four miles from Microsoft headquarters in Redmond, WA) and can be reached at (425) 895-0550. To reach Prime Recognition on-line, visit: http://www.primerecognition.com

PrimeOCR, PrimeView and PrimeVerify are trademarks of Prime Recognition. Windows is a trademark of Microsoft Corp. All other products are trademarks or registered trademarks of their respective companies.

For more information contact: marketing@primerecognition.com

Prime Recognition
21827 NE 137th St.
Woodinville, WA 98072
425-895-0550
425-895-9580 Fax
http://www.primerecognition.com