OREANDA-NEWS. July 17, 2012. ABBYY®, a leading provider of document recognition, data capture, linguistic technologies and services, announced the release of ABBYY Recognition Server 3.5, an intelligent server-based software for automating document capture across an organization. The new release features the latest ABBYY innovations in the area of optical character recognition (OCR) technology delivering powerful Arabic OCR and higher accuracy of document analysis, and offers enhanced scanning functionality.

ABBYY Recognition Server converts large volumes of paper documents or images into fully searchable electronic text suitable for document archiving, e-discovery, enterprise search and other important business processes. Designed to streamline document processing in organizations, the product encompasses all the essential parts of the document capture process: scanning, recognition, document separation, classification, indexing and delivery. It enables automated, unattended document processing which can be centrally managed and accessed from multiple points within the organization or remotely. It can also connect with a variety of back-end systems and third-party applications with integration via Scripts, XML tickets, a Web-service API, or a COM-based API.

One of the key technology innovations of Recognition Server 3.5 includes the Arabic OCR functionality developed by ABBYY — a result of several years of research and development. The software can now process Arabic texts with high accuracy, overcoming technical challenges previously associated with recognition of Arabic characters. With the new addition, ABBYY Recognition Server can address document processing needs of a whole new set of corporate customers, and becomes a truly international product. Based on the latest OCR technology, version 3.5 also delivers up to 17 percent higher accuracy in detecting blocks during document analysis and 33 percent higher accuracy of table element detection and recognition.

ABBYY Recognition Server 3.5 also delivers enhanced customization capabilities such as the new Scanning Station with scripting support. It now allows administrators to easily automate scanning operations by creating user-specific processing scenarios, such as placing scanned pages in the correct order after duplex scanning, auto-split of large-size pages after scanning, saving pages to two destinations simultaneously, and more.

Additionally, it offers the enhanced scenario for processing document images sent by e-mail, allowing users to preserve the initial subject of the e-mail or specify a custom one. This functionality is crucial for business processes where e-mail subject lines define further processing steps for the recognized document.

Lastly, the Portuguese (Brazil) and Czech languages were added to the existing set of supported user interface languages.