ABBYY has announced the latest release of FlexiCapture 10, its software for data and document capture, now provides access to all document scanning and verification functions via Web interfaces, for deployment in SaaS and cloud environments.

It also allows companies to quickly implement distributed document capture processes, saving resources typically spent on local software installation and maintenance. ABBYY FlexiCapture 10 now offers a complete set of Web-based operator stations including a new Web Capture Station that combines both scanning and verification functions in a single interface.

Additionally, it offers a range of enhancements, such as content-based auto-learning technology for document classification and extended verification functions.

ABBYY FlexiCapture 10 provides intelligent technologies for automating the process of extracting data from a variety of documents, including machine-readable forms, semi-structured and unstructured documents. It can process multiple types of documents in a single stream and scales from a standalone application to enterprise-level projects with distributed processing.

Additionally, it provides highly intuitive tools for setting-up document capture processes, enabling users to train the system with simple point-and-click operations.

The product now includes a dedicated Web Scanning Station, a Web Verification Station and a brand new Web Capture Station, which combines both of those functionalities in a single interface.

Web-based stations are designed to address the needs of organisations with a distributed network of offices or points of service, such as insurance companies, banks, healthcare providers and travel agencies.

FlexiCapture 10 now supports "secure" verification designed for better protection of sensitive information in documents. It can divide all data contained in one document into several groups and forward them separately to different operators for verification. For example, one operator checks "sums" while another is checking "names" from the same documents, yet they are unable to interconnect those pieces of data.

Since its initial release, FlexiCapture can be trained to automatically distinguish documents of different types based on a set of sample images. With the latest product release, its auto-learning technology has been further enhanced to support both image-based and content-based approaches to the document classification.

In the image-based classification mode, the system can quickly process similar looking documents, such as forms or invoices from the same vendor, and allows users to achieve faster data extraction. The new content-based approach helps classify and process unstructured documents that may have very different layouts within the same document type (e.g. contracts, letters, etc.). Content-based classification applies full-text analysis of documents looking for words and phrases that are considered distinctive for certain document types.

ABBYY FlexiCapture 10 now supports 13 interface languages including English, French, German, Hungarian, Chinese (simplified characters), Czech, Italian, Korean, Polish, Portuguese (Brazilian), Russian, Serbian, and Spanish.