AWS Receipt Processing Lambda Function

This project uses AWS Textract, DynamoDB, SES, and S3 to automatically process receipt images uploaded to an S3 bucket. When a new receipt is uploaded, this Lambda function:

1. Extracts structured data using Amazon Textract
2. Saves the extracted data to DynamoDB
3. Sends an HTML-formatted email notification using Amazon SES

Architecture Overview

Features

• Triggered automatically by new S3 uploads
• Intelligent field extraction using analyze_expense API from Textract
• Saves full receipt metadata and line items to DynamoDB
• HTML email notifications with vendor, total, date, and item breakdown
• Resilient error handling with clear logs
• Easy to configure using environment variables

Technologies Used

• AWS Lambda
• Amazon Textract
• Amazon S3
• Amazon DynamoDB
• Amazon SES

Setup Instructions

1. Prerequisites

• AWS account
• Verified email in Amazon SES (for both sender and recipient)
• IAM Role with permissions for:
  • Textract: textract:AnalyzeExpense
  • S3: s3:GetObject, s3:HeadObject
  • DynamoDB: dynamodb:PutItem
  • SES: ses:SendEmail

2. Environment Variables

Configure the following Lambda environment variables:

Variable	Description
DYNAMODB_TABLE	Name of the DynamoDB table (e.g. Receipts)
SES_SENDER_EMAIL	Verified SES sender address
SES_RECIPIENT_EMAIL	Destination email for receipt summaries

Repository Structure

Automated-Receipt-Processing-System
├── lambda.py
├── Lambda Function Explanation.pdf
├── README.md
└── screenshots/

Screenshots