In general ABBYY technology is known for its high quality OCR results, but there are scenarios where users/operators are involved. This can be the case when:
- Text in documents, e.g contracts should be checked before exported to e.g. Microsoft Word
- Extracted data has to be 100% correct
- Documents with images should be converted e.g. to ePub - Operators can check that all images and tables are detected perfectly
Developers have full access to the internal recognition results. FineReader Engine provides basic information like: the character coordinates, but also very advanced attributes, like: font and formatting information and word and character recognition hypotheses.
The information is available via API and XML Export, so that they can be used for automated correction. For simplified, user driven correction and verification, starting from FineReader Engine 9.0 (Windows only) visual components as ActiveX controls are part of the SDK. So layout analysis results and uncertainly recognized characters can be changed, but also the page order within a document.
Example of Visual Components:
Please sign in to leave a comment.