Back to Newsroom

ABBYY FineReader Engine 9.0 Software Toolkit Brings New Levels of Accuracy and Performance to Multipage Document Conversion

October 21, 2008

ABBYY announced its FineReader® Engine® 9.0 for Windows®, the newest software development kit (SDK) to integrate ABBYY’s multilingual and multiplatform recognition technologies.

New Version Combines Breakthrough Technology to Accurately Re-Create Document Logical Structure with Outstanding Performance on Multi-Core Processors

ABBYY, a leading provider of document recognition, data capture and linguistic software, today announced its FineReader® Engine® 9.0 for Windows®, the newest software development kit (SDK) to integrate ABBYY’s multilingual and multiplatform recognition technologies. Available today, FineReader Engine 9.0 offers breakthrough technology to deliver new levels of recognition accuracy and performance, particularly in processing multipage documents. An enhanced core technology platform combined with additional improvements, such as significantly increased scalability, new technology for Asian language recognition and ready-made interface tools, make ABBYY FineReader Engine 9.0 a definitive solution for integrating highly accurate, scalable and efficient recognition technologies.

ABBYY FineReader Engine 9.0 marks a significant step forward in delivering high quality document recognition and PDF conversion of multipage documents. New Adaptive Document Recognition Technology (ADRT™) not only provides highly accurate reproduction of text and standard formatting, but precisely re-creates text flow, logical integrity and formatting consistency across an entire document. As a result, formatting attributes of the original - such as headers, footers, page numbers, footnotes, columns, tables, signatures, fonts and styles - are accurately reconstructed in native Microsoft Office formatting.

Andrey Isaev, director of the technology products department at ABBYY, explained how ADRT moves document recognition from a page-based to a document-based approach to accurately re-create the logical structure and formatting across multiple pages. “Unlike traditional recognition which processes a document as just a combination of single pages, ADRT processes the document as a single entity,” said Isaev. “It analyzes cumulative data from all pages of the document to make accurate hypotheses about its general structure and formatting attributes. The result is output of documents which are truly editable and reusable.”

FineReader Engine 9.0 also delivers significant improvements in performance, accuracy and additional functionality:

High Scalability on Multi-Core Processors

In addition to the enhanced multipage document recognition capabilities, FineReader Engine 9.0 delivers outstanding performance on multi-core processors. New extended CPU core support allows FineReader Engine to effectively distribute document pages among CPU cores. The SDK is designed to scale depending on the number of CPU cores, providing a significant increase in processing speed with each new core added to the system. For example, it can deliver up to 90 percent* increase on dual-core, up to 250 percent* increase on quad-core processors, and so on.

New Recognition Technology for Chinese and Japanese

Featuring a completely new and powerful recognition technology for Asian languages, FineReader Engine 9.0 delivers a major breakthrough in recognition performance and accuracy for Chinese and Japanese, two of the most complex languages for recognition and conversion. A new intelligent language detection algorithm delivers exceptional results in processing multilingual documents, and enables the solution to successfully convert documents with various combinations of ideographic (e.g. Asian) and alphabetical languages (e.g. European) such as Chinese and French or Japanese and German.

“With the continual growth of the Asian marketplace, languages such as Chinese are becoming more and more predominant in a variety of business documents and communications,” explained Isaev. “Yet these types of languages with thousands of characters have always presented an ongoing challenge for recognition that needed to be addressed. To this end, ABBYY has invested significant resources in technology development and we are proud to announce powerful enhancements in this area.”

Ready-to-Use Visual Components for Easy Implementation of OCR Functions

New, ready-to-use Visual Components provide a set of graphical user interface elements which integrators can easily take advantage of when incorporating recognition capabilities into their applications. Available in 24 interface languages, Visual Components include intuitive interfaces for scanning/opening document images, automatic or manual zoning of the image, text editing, text validation and more. All the components leverage ABBYY’s extensive experience in developing and marketing end-user applications such as the award-winning ABBYY FineReader optical character recognition (OCR) application.

Innovative Compression Technology for PDF and PDF/A Export

ABBYY FineReader Engine 9.0 delivers an innovative Mixed Raster Content (MRC) compression technology for PDF and PDF/A. The MRC compression for PDF is capable of reducing the size of output file by up to 8-10 times* over the usual JPEG compression. The technology identifies different layers on PDFs and effectively compresses them without loss of visual quality. Support for highly compressed searchable PDF and PDF/A is crucial for developing today’s document archiving and storage applications.

Next-generation ABBYY Camera OCR Technology

The new SDK version includes ABBYY’s advanced innovations to identify photographed images and intelligently recognize text on them. ABBYY Camera OCR provides enhanced image pre-processing functions including image resolution identification, as well as correction of curved lines, skewing and other distortions typically found on photographed images. 

Other enhancements of the FineReader Engine 9.0 include:

  • New export to XML-based MS Office 2007 formats such as DOCX, XLSX, and PPTX;
  • Support for new JBIG2 image format;
  • Enhanced language support - The SDK adds support for 21 new languages for ICR (recognition of hand printed characters) providing support for a total of 113 languages for ICR and 195 languages for OCR (recognition of printed characters).  FineReader Engine 9.0 also includes accuracy enhancements for Hebrew language recognition.

About ABBYY Recognition Technologies and ABBYY Multiplatform Toolkits

ABBYY FineReader Engine is a powerful SDK that gives developers and integrators the tools they require to integrate a variety of recognition technologies into their software applications. The ABBYY recognition platform delivers award-winning OCR, intelligent character recognition (ICR), barcode, checkmark, field-level/zonal recognition and PDF conversion for transforming scanned documents and image files into searchable, editable and manageable text files. Already providing the recognition component for leading enterprise document management, workflow and archiving systems from leading providers, ABBYY FineReader Engine toolkits are available to support multiple operating environments including Windows, Linux®, FreeBSD® and Mac OS®. ABBYY also offers an OS-independent, small code-sized development toolkit specially designed for integrating OCR and business card reading functions into any mobile application and operating system including Windows Mobile, Symbian and Linux.

Availability and Pricing

ABBYY FineReader Engine 9.0 is available immediately worldwide via a flexible, modular licensing policy. Developers may select the best combination of tools and pricing options for their projects. Pricing varies according to the number of CPU cores, processing stations and number of pages processed. For information on licensing models and pricing, contact your local ABBYY office or ABBYY partner. 

A special time-limited trial version is also available for testing. For more information about the product, visit *Numbers quoted are based on internal ABBYY testing.

Connect with us