2.7.0

In this Version, PDF processing has been improved in the OCR engine. In addition, the licensing backend has been improved, incorporating high availability, and the facephi-recognition library.

Added

• PDF: the default threshold is reduced to 0.5 to improve text detection.

• PDF: the optional parameter is added pageNums. It is a list of page numbers to be processed. Numbering starts at 1. If not specified, all pages will be processed.

• PDF: the "page not parsed" errors are ignored on the 2nd page and beyond.

• PDF: the output is added warnings. This field contains information about pages that have not been processed. In the future, it could be expanded to include other types of information.

• text-interpreter update:

  • PDF: the model fields are included even if they are not detected or are empty.

• document-reader-sdk-templates update

  • PDF: the IVA.json and RENTA.json models correctly process multi-word fields.

• Licensing backend Version 2.0.0 update.

  • High availability is added.

• Update of facephi-recognition v3.21.0