ORPALIS launches a key-value pair data extractor in its OCR SDK

ORPALIS Imaging Technologies

ORPALIS is pleased to introduce KVP extraction in its OCR engine.

Extracting key-value pairs is at the heart of intelligent document processing systems.

Around 90% of all documents used by a business or organization are unstructured.

Therefore, extracting information from invoices, contracts, forms, bank statements or emails can be tedious. It is also difficult to index and reuse this information elsewhere.

A KVP engine automatically extracts meaningful information from unstructured and semi-structured documents.

ORPALIS is pleased to introduce KVP extraction in its OCR engine.

Like other OCR technologies developed internally by the company (MICR, MRZ, OMR, contextual OCR, etc.), the KVP extractor benefits from a hybrid approach that includes heuristics, mathematics and ML capabilities.

The engine is based on an understanding of adaptive layout and the same underlying element techniques as NLP technologies.

The KVP extraction engine automatically adapts to the document and searches for the right approach, making the best use of available resources.

This approach gives excellent results on the usual weaknesses of traditional OCR and pure Machine Learning engines, in particular with:

  • Text recognition in documents with a lot of noise,
  • Dotted,
  • Touching & broken characters,
  • Text on colored background,
  • underlined text,
  • biased text,
  • Text in graphs and tables.

In addition to the key and the value, the ORPALIS engine also provides the type (nature of the content) and the precision (level of confidence).

The KVP extractor is available with the latest download of GdPicture.NET and DocuVieware SDKs.

More information on the GdPicture.NET website.



ORPALIS is a publisher of imaging software, PDF processing tools and large-scale document flow management solutions for professionals around the world.

In 2022, the French company joined PSPDFKit, the leading document processing and manipulation platform for developers and businesses.

ORPALIS sits on the Board of Directors of the PDF Association.

For more information, visit http://www.orpalis.com.

Share the article on social networks or by e-mail:

Comments are closed.