Amazon Textract

Automatically detects and extracts printed text and handwriting from scanned documents and images using OCR and machine learning.

Extracts structured data such as key-value pairs from forms, maintaining the relationship between fields and their values.

Identifies and extracts tabular data organized in rows and columns from documents for structured data processing.

Allows users to specify natural language questions to extract specific information from documents regardless of layout variations.

Enables customization of query-based extraction for improved accuracy on business-specific documents by training adapters.

Detects handwritten and electronic signatures and initials on documents and images.

Extracts and understands context from identity documents like U.S. passports and driver’s licenses without templates.

Processes documents at scale with data privacy, encryption, and compliance standards, supporting multiple file formats and languages.