Amazon Textract – Analyze document action

The Analyze document action uses Amazon Textract to extract structured data, such as tables, forms, queries, and signatures, from documents stored in an Amazon S3 bucket.

Input

Input field	Description
S3 bucket	Enter the name of the S3 bucket containing the document.
S3 object key	Enter the path or key to the object within the S3 bucket.
Feature types	Select one or more feature types to detect. You can choose from FORMS, TABLES, QUERIES, or SIGNATURES.

Output

Output field	Description
Document metadata	Metadata about the input document.
Blocks	Array of recognized content blocks including key-value pairs, text, and page.
Warnings	Any warnings returned during document analysis.
Analyze document model version	The version of the Textract model used for the analysis.
Human loop activation output	Information about whether a human review was triggered.