On this page
Amazon Textract: document processing features and setup
An AWS service for extracting text and structure from scanned documents. It is designed for developers building document workflows who need information from tables, forms, and fields beyond continuous text.
Tool by the Otomatzia editorial team · source content checked on
Overview
Textract is one component in a broader AWS workflow, so plan where the document comes from and where the output goes. An extracted table may need mapping to fields in a business system, and a form field should remain linked to its label. Choose the extraction capability based on the information you need, rather than assuming every interface provides the same data structure for every document.
Features
- Recognizes printed text and handwriting in supported documents
- Extracts tables and form data
- Ready-made capabilities and customization for document scenarios
Business uses
- Convert a document table into a data structure for review
- Prepare information from a scanned form for further processing in a system
Getting started
- Choose a document type and extraction interface, then run an example in a restricted AWS environment.
- Run a form and a table as separate examples. Check the output structure, including an empty row and a missing field, before writing the mapping to the receiving system.
- Try a document in the target language and at the expected scan quality. Then check that access permissions for the file and result are limited to the components required by the AWS workflow.
What to check
- Check the list of supported languages, especially before using documents in Hebrew.
- Review the processing region, storage permissions, and costs for the selected extraction operation.
Sources to check next
Vendor features and prices change. Confirm details on the linked official pages before you rely on them.