The latest version of ABBYY Recognition Server 4.0 is using the lasted ABBYY OCR technology.
This implementation leads to a range of new features:
Digital archives and document libraries can be recreated in searchable format automatically, without copying files into Hot folders. Images will be converted into searchable PDFs, while documents created on a PC, such as Word or Excel documents, can be moved directly to the newly created library that will mirror the structure of the source library. Documents that were added later will be regularly detected and processed. This way any library with different set of documents (scans, Word documents, PDFs) can be changed to a library with a fully searchable content.
Existing scanned PDFs and image documents can be converted to searchable PDFs directly within the SharePoint library. The SharePoint Library can be selected as source of documents and all detected image files can be automatically turned into searchable PDFs. Scanned PDFs can be enhanced by a text layer and turned into searchable PDFs while PDFs that already have a text layer (created on PC) can be skipped.
PDFs that are not searchable (scanned PDFs or PDFs with an insufficient text layer quality) can be detected and enhanced with a high quality text layer. Existing bookmarks and annotations can be kept. New PDF/A formats have been added. Recognition Server now supports following PDF/A formats: PDF/A-1a, PDF/A-1b, PDF/A-2a, PDF/A-2b, PDF/A-2u
New SQLite database and improved fault tolerance mechanisms provide robust and consistent handling of high volumes of documents for multiple clients. Adminstration is easy with enhanced logging and user management via Active Directory.
Faster Arabic recognition and recognition of technical drawings with text printed sideways. ABBYY Recognition Server delivers up to 20 percent* faster processing of Arabic texts and up to 10 percent1) more accurate results on office documents and books in Ara
To speed up processing the Operator of the Verification Station can select urgently needed documents in the document queue and verify them first. For verifiying extensive documents (like books) the results can now be saved even if the verification of a document is not yet finished (manually or automatically).
Indexing value fields can now be imported from external sources such as databases or .csv files. Individual documents can be selected from a document Queue on the Indexing Station and indexed with higher priority. The indexing results can be saved (manually or automatically) to prevent data loss.