Document Intelligence helps you to quickly and accurately classify and extract information from documents using artificial intelligence (AI).

Document Intelligence overview

Document Intelligence uses optical character recognition (OCR) together with AI models to detect and extract information from documents with various structures and formatting. This enables you to accurately extract information and automate document processing.

Document Intelligence workflow

With Document Intelligence (DocIntel) you can process single or multi-page documents in JPEG, PNG, or PDF formats. You can process documents that contain typed text such as forms, invoices, identity documents, and more.

The following diagram shows how document extraction works in Document Intelligence.

Figure 1. Document Intelligence flow
Diagram showing how Document Intelligence activities train the AI models.

In this workflow:

  1. A document is uploaded for processing in a document task.
  2. DocIntel extracts the data from the document using OCR and AI models.
  3. The user provides input to validate or correct the DocIntel recommendations.
  4. The models are updated and trained to provide more accurate results.

Document Intelligence benefits

Figure 2. Benefits of Document Intelligence
Diagram showing the phased approach to automation using Document Intelligence.
Document Intelligence provides the following benefits.