Adaptive Document Recognition Technology (ADRT)

New recognition technology based on an innovative set of document synthesis algorithms. Automatically builds a logical model of the document structure and identifies its elements.

• Formatting purpose of elements like headers and footers, footnotes, page numbers etc.
• Order in which document should be read
• How document elements should be reconstructed

ADRT automatically locates, determines and reconstructs various structural parts of the document and formatting of the elements. Documents generated by ADRT have consistent formatting across all pages of a document since they are processed as a unit.

Multi-Page Processing through new Document specific API

The new API objects allow you to set up the parameters of page and document synthesis separately. The pages will be processed as a logical unit and not isolated from each other. This approach preserves the logical organisation of the document, retaining not only the original text and columns, but also fonts, styles, etc.

Multi CPU / Multi Core Recognition Architecture

The new Utilises all CPU cores during analysis and recognition of multi page documents. This function helps to automatically combine and execute step of distributing pages, as well as coordinating recognition and the synthesis.

New image pre-processing capatibilities

• Detection of an image rotation up to 20 degrees
• Deskew by horizontal and vertical pairs of black squares
• Deskew by horizontal and vertical lines
• Deskew by horizontal and vertical lines of text

Visual Components

Visual Components offer a set of graphical user interface components which developers can take advantage of when creating their own applications. Visual Components were created using ABBYY’s expertise in end-user applications and are based on interfaces created for our award-winning FineReader Professional Edition application.

• Scan Interface
  Provides an interface for accessing TWAIN-based scanners. Allows users to select key scanning settings such as resolution, paper settings, etc.
• Document Viewer A visual overview of documents images and status of recognition processing. Available in a thumbnail view of each page, or a detailed view in table format.
• Image Viewer
  A full image view of a document page- it can be processed with specialised tools for creating, editing and changing the settings of recognition blocks. Tools include for example:
  • Rotating, cropping and splitting images.
  • Drawing recognition areas or selecting block types – text, image, table, barcode.
  • Buttons of the toolbar can be displayed or hidden via code, customised buttons are possible
• Text Editor
  Provides option to highlight uncertain characters and basic text formatting tools. Developers can control the text area display, available buttons and the user actions, like the selected text areas
• Text Validator
  An easy-to-use Wizard-like tool to check recognised characters marked as “uncertain.” Includes an integrated function to check spelling and provides a “zoom” view of key portions of text to be checked. Developers can control the behaviour of this component.

Improved Asian Language OCR Support

New ABBYY technology for Chinese and Japanese languages. Functionality enhancements include improved accuracy, faster processing and the ability to combine these ideographic languages with other languages (not only with English as it was in the previous version).

New Export Formats

• XML-based Office 2007 File Formats
  This feature makes it possible to export to new XML-based Microsoft Office formats: DOCX, XLSX and PPTX.
• MRC (Mixed Raster Content) Compression for PDF and PDF/A
  Export to PDF with MRC compression. Supports the ability to set the MRC compression level. The parameters of compression for background, colour - and text mask and change background and text colour.

New and extended Licensing Options

• Extended CPU Core Support
  Out-of-the box support for multi core CPUs. New licensing scheme allows an unlimited number of cores with page limited licences.
• CPU core based licences
  New offer of licences without a page counter - instead pricing is on the maximum number of CPU cores that can be used.
• Maximum Speed Limitation*
  This new option sets maximum recognition speed limitation on characters per second.