IT and Culture: ABBYY’s technologies facilitate digitization of Europe’s cultural heritage

ABBYY, a leading provider of document recognition, document capture, and linguistic technologies and services, has announced the first results of its participation in the European Commission's IMProving ACcess to Text (IMPACT) project. European-wide research project aims to transform Europe's printed heritage into digitally available resources. The IMPACT consortium brings together twenty-six European national and regional libraries, research institutions and technology partners sharing their knowledge and best practice, and developing innovative tools to enhance the capabilities of OCR engines.

Since 2008, ABBYY has played a key role in the IMPACT project by providing state-of-the-art OCR technology and expertise for digitising historic fonts and extracting text from old documents. ABBYY provides Optical Character Recognition (OCR) software and expertise in recognising historic fonts for this European-wide research project. Using historically relevant samples collected by leading European libraries, ABBYY has delivered new technology advancements in image pre-processing and analysis of document layouts for better character recognition results. Enriched by the experience gained in the four-year project, ABBYY now offers products equipped with the most advanced text recognition technology for printed historic documents. The latest enhancements to ABBYY recognition technologies are now available commercially for use by institutions and organisations around the world. ABBYY FineReader Engine SDK and ABBYY Recognition Server 3.0 deliver some of ABBYY's recent improvements in the areas of recognition of specialised texts and documents.

ABBYY research and development teams have been working closely with key members of the IMPACT project teams to address key technological issues faced by large library digitisation projects. ABBYY's OCR technology is particularly suited to recognise text on images of documents in various historic typefaces making it possible for those resources to be fully digitised, searchable and integrated with systems used for lexicon building.