Machine Learning

Automatic Data Extraction from Documents with Amazon Textract - OCR, Table Analysis, and Form Recognition

Learn how to extract text from documents, analyze table structures, and extract key-value pairs from forms using Textract.

約 3 分で読めます

Overview of Textract

Textract is an OCR service that automatically extracts text, tables, and form data from documents, supporting PDFs up to 3,000 pages and images up to 10 MB. While traditional OCR only recognizes text positions and characters, Textract understands table row-column structures and the relationships between form labels and values. The Queries feature extracts answers to specific questions from documents, and AnalyzeExpense structures invoice line items.

Table Analysis and Queries

The AnalyzeDocument API's Tables feature recognizes table rows and columns, returning cell contents as structured data. It correctly handles merged cells and header rows. The Forms feature automatically pairs form labels (such as "Name," "Address," "Phone Number") with their corresponding values. Queries lets you ask natural language questions about a document, extracting answers to questions like "What is the patient's name?" or "What is the total amount?" AnalyzeExpense specializes in receipts and invoices, automatically classifying vendor names, dates, total amounts, tax amounts, and line items.

AnalyzeExpense and Lending

The AnalyzeExpense API specializes in invoices and receipts, extracting vendor names, invoice dates, total amounts, and line items (item names, quantities, unit prices) as structured data. It handles handwritten receipts and multi-page invoices, making it ideal for automating expense reporting. The AnalyzeLending API specializes in loan documents such as mortgage applications, automatically classifying document types like application forms, income statements, and property appraisals before extracting fields from each. The asynchronous API (StartDocumentAnalysis) enables batch processing of large document volumes with results output to S3. Integration with A2I (Augmented AI) lets you route low-confidence extraction results to human review workflows. For a deeper understanding of OCR theory and implementation, specialized books on Amazon are a great resource.

Optimizing Textract Costs

Textract pricing is based on the API type and number of pages. DetectDocumentText (text extraction only) costs approximately $1.50 per 1,000 pages, AnalyzeDocument (table and form analysis) costs approximately $15, and Queries cost approximately $0.015 per query. AnalyzeExpense costs approximately $10 per 1,000 pages. Use DetectDocumentText when text extraction alone is sufficient, and only use AnalyzeDocument when table or form structure analysis is needed to optimize costs. Pre-processing documents to exclude unnecessary pages (blank pages, cover pages) reduces the number of pages processed. The asynchronous batch API has the same pricing as the real-time API but is effective for avoiding throttling during large-scale processing.

Summary

Textract is an advanced OCR service that goes beyond text extraction to understand table structures and form key-value pairs. Queries extract answers to specific questions from documents, and AnalyzeExpense structures invoice line items. AnalyzeLending automates classification and extraction of loan documents, and A2I integration routes low-confidence results to human review workflows.

Related Services

Amazon TextractAn OCR service that automatically extracts text, tables, and form data from documentsAmazon S3A scalable object storage serviceAWS LambdaA serverless compute service that runs code without server management

Related Articles

Implementing Natural Language Processing with Amazon Comprehend - Sentiment Analysis and Entity ExtractionLearn about sentiment analysis, entity extraction, and building custom classification models with Comprehend.Automating Document Processing with Amazon Textract - From OCR to Form and Table ExtractionGo beyond OCR with structural recognition of forms and tables to automatically extract data from invoices, receipts, and identity documents. Also covers integrating human review with A2I.Practicing Infrastructure as Code with AWS CloudFormation - Template Design and Stack ManagementDefine infrastructure with templates and preview impact with change sets before deployment. Detect configuration drift with drift detection and deploy across your entire organization with StackSets.

More on This Topic

GPU-Based Machine Learning Training with AWS Batch - Cost-Efficient Large-Scale TrainingRun GPU training with your existing Docker containers, and cut costs by up to 90% using Spot Instances and checkpointing. Includes guidance on when to choose Batch over SageMaker.Using Claude on Amazon Bedrock - Model Selection, Prompt Design, and Cost OptimizationCompares the Anthropic Claude models available on Amazon Bedrock, provides model selection guidelines by use case, and covers prompt design best practices and cost optimization.Building RAG Applications with Amazon Bedrock Knowledge Bases - Implementing Retrieval-Augmented GenerationAutomatically index documents on S3 and unify search and generation with the RetrieveAndGenerate API. Covers chunking strategy selection and safety enforcement with Guardrails.Getting Started with Quantum Computing on Amazon Braket - Designing and Simulating Quantum CircuitsPrototype for free with local simulators, then run quantum circuits on IonQ and Rigetti hardware. Covers implementing VQE and QAOA with hybrid jobs.Privacy-Preserving ML with AWS Clean Rooms ML - Build Models Without Sharing DataLearn how to build lookalike models with Clean Rooms ML, apply differential privacy, and leverage the results for ad targeting.Implementing Natural Language Processing with Amazon Comprehend - Sentiment Analysis and Entity ExtractionLearn about sentiment analysis, entity extraction, and building custom classification models with Comprehend.Building Conversational Bots - Natural Conversation Interfaces with Amazon Lex and PollyLearn how to build conversational bots using Amazon Lex and Amazon Polly.Demand Forecasting - Predicting the Future from Time Series Data with Amazon ForecastInput historical time series data and related variables to automatically build ML-based demand forecasting models. This guide covers forecast accuracy evaluation metrics and patterns for leveraging forecast results through S3 and QuickSight integration.

Similar Articles and Services

Automating Document Processing with Amazon Textract - From OCR to Form and Table ExtractionGo beyond OCR with structural recognition of forms and tables to automatically extract data from invoices, receipts, and identity documents. Also covers integrating human review with A2I.Document Text Extraction - Intelligent Document Processing with Amazon TextractLearn how to automatically extract text, tables, and form data from documents with Amazon Textract, and build natural language processing pipelines by integrating with Amazon Comprehend. This article covers automation patterns for invoice processing and contract analysis.Amazon TextractA machine learning-based document analysis service that automatically extracts text, handwriting, tables, and form key-value pairs from documentsImplementing Natural Language Processing with Amazon Comprehend - Sentiment Analysis and Entity ExtractionLearn about sentiment analysis, entity extraction, and building custom classification models with Comprehend.