Receipt Extractor

A Python tool for extracting and processing receipts from images. This project provides both a command-line utility and a REST API for background removal and processing of receipt images.

Features

• Remove backgrounds from receipt images
• Crop to content boundaries
• Enhance images for OCR processing
• Process single images or batches
• Support for both file-based and base64 image processing
• REST API with FastAPI

Installation

# Clone the repository
git clone https://github.com/David031/receipt-extractor.git
cd receipt-extractor

# Install dependencies
pip install -r requirements.txt

Usage

Command-line Tool

# Process a single image
python run.py input.jpg output.jpg

# Process with OCR enhancement
python run.py input.jpg output.jpg --enhance

# Process multiple images in a directory
python run.py input_folder/ output_folder/ --batch

API Server

# Start the API server
python run_api.py

The API will be available at http://localhost:8000

API Endpoints

• GET / - Root endpoint with API information
• POST /remove-background - Remove background from an uploaded image
• POST /remove-background-base64 - Remove background from a base64 encoded image
• POST /batch-process - Process multiple uploaded images
• POST /batch-process-base64 - Process multiple base64 encoded images
• GET /health - Health check endpoint
• GET /metrics - Performance metrics endpoint

Dependencies

• rembg
• opencv-python
• numpy
• Pillow
• FastAPI
• scikit-image
• PyYAML
• requests
• psutil