ABBYY Launches First OCR Product to Deliver Accurate Recognition of Logical Document Structure
Released on: December 26, 2007, 5:58 am
Press Release Author: ABBYY
Industry: Software
Press Release Summary: Create editable & searchable files from scanned documents, PDF and camera images
Press Release Body: Moscow, Russia (October 1, 2007) - ABBYY, a leading provider of document recognition, data capture and linguistic technologies, today announced version 9.0 of the ABBYY FineReader document and PDF conversion software. Based on a breakthrough new recognition platform with Adaptive Document Recognition Technology (ADRTT), FineReader 9.0 is the first optical character recognition (OCR) product intelligent enough to automatically deliver editable files in native document formatting. The new version offers up to 35 percent* accuracy improvements, particularly with documents containing tables and images from digital cameras. FineReader 9.0 combines enhanced accuracy in the reproduction of document layouts with new tools, extended document format support, and an innovative interface redesign to provide increased user productivity with less steps and effort spent on OCR tasks than ever before. ABBYY FineReader provides comprehensive and easy-to-use document conversion capabilities which allow users to readily transform document images and PDF files into editable and searchable electronic files. A cost-effective and time-saving alternative to manually retyping documents, FineReader delivers precision accuracy in converting almost any type of document, including tables and spreadsheets, magazine articles, contracts, and even faxes. ABBYY's unique ADRT recognizes not just the text and basic layout, but the logical structure of a document across multiple pages, producing an improvement in layout retention by up to 32 percent*. Because it understands a document as a whole, FineReader is very intelligent in saving and exporting OCR results. It can determine how certain formatting elements should be treated when sending to Microsoft® Word and Microsoft Excel®. As a result, the headers and footers, page numbering, footnotes, fonts and styles of the original are retained. For example, footnotes linked with corresponding text on the page, image captions, graphics and tables are automatically grouped with the appropriate object type. Headers and footers can be directly edited or even removed using the Microsoft Word header and footer tools. Text flows are maintained across multiple pages. The new 9.0 recognition platform also recognizes and retains a variety of additional formatting elements including line numbering, signatures, and stamps found in legal and other documents. "Through each step of OCR development, we have been able to reproduce an increasing amount of document parts automatically - from individual letters and text, to formatting elements such as columns and tables," explained Sergey Popov, Chief Product Manager at ABBYY. "Now with the ABBYY ADRT, we can analyze not just the separate formatting elements of a single page, but also the formatting and logical text flow across a document with multiple pages, and recreate the document as a logical whole."