The provided recognition and processing engines fully automate document classification and data extraction for known forms. The recognition engine creates unique XML data for each known form which is used to identify any form requiring recognition. The processing engine uses templates consisting of predefined fields for each known form. These fields can be defined using the Forms application included with the SDK, or developers can create their own template editor using the toolkit. Supported field types include machine and hand written characters, check boxes, filled bubbles, cross marks and barcodes.
For maximum performance, forms recognition and OCR high level interfaces utilize an intelligent multi-threaded algorithm during recognition and processing and are supported in both 32 and 64 bit development.
In addition to saving the recognized data in various structured output formats such as delimited text, XML or database fields, a recognized free form document can be converted to searchable PDF for enterprise data mining.
