xsukax Offline English OCR

A privacy-focused, fully offline Optical Character Recognition (OCR) solution for extracting text from images and documents without sending your data to external servers.

Github Repo

Project Overview

xsukax Offline English OCR is a robust, self-contained OCR system designed to process various document formats while maintaining complete data privacy. The application leverages multiple OCR engines (EasyOCR and Tesseract) to maximize text extraction accuracy from images (JPG, PNG, BMP, TIFF, WEBP), PDF documents, and Microsoft Word files (DOCX, DOC).

The system provides both a RESTful API built with FastAPI and a modern web interface for easy interaction, making it suitable for both developers and end-users who need reliable text extraction without compromising sensitive information.

Security and Privacy Benefits

In an era of increasing data surveillance and privacy concerns, xsukax Offline English OCR offers significant security advantages:

• Complete Data Sovereignty: All processing occurs locally on your hardware – no data ever leaves your environment
• No Internet Dependency: Once installed, the system operates entirely offline with no external API calls
• Temporary File Management: Automatically cleans up processed files with secure deletion practices
• No Telemetry or Tracking: The application contains no analytics, tracking, or data collection mechanisms
• Local Model Storage: All OCR models are downloaded and stored locally during installation
• Transparent Code: Open-source architecture allows full inspection of all processing logic

Features and Advantages

• Dual-Engine OCR: Combines EasyOCR and Tesseract engines for improved accuracy across different document types
• Batch Processing: Support for processing up to 10 files simultaneously
• Document Format Support: Handles images, PDFs, and Word documents with a unified interface
• Image Enhancement: Automatic pre-processing to improve OCR accuracy on challenging documents
• REST API: Developer-friendly API for integration with other applications
• Modern Web Interface: Intuitive user interface for non-technical users
• Cross-Platform Compatibility: Works on Windows, macOS, and Linux systems
• Configurable Processing: Adjustable parameters for different OCR scenarios

Installation Instructions

Prerequisites

• Python 3.8 or higher
• 4GB+ RAM recommended (for OCR processing)
• 2GB+ disk space (for model storage)

Automated Installation

1. Clone or download the project files
2. Run the installation script:

python install_deps.py

The installation script will:

• Create necessary directories structure
• Install system dependencies (Tesseract OCR and Poppler for PDF processing)
• Install Python package dependencies
• Download and configure OCR models
• Create configuration file
• Verify the installation

Manual Installation (Alternative)

For systems where the automated script doesn’t work:

1. Install system dependencies:

Ubuntu/Debian:

sudo apt-get update
sudo apt-get install -y tesseract-ocr tesseract-ocr-eng libtesseract-dev poppler-utils

macOS:

brew install tesseract poppler

Windows: Download and install:

• Tesseract OCR
• Poppler for Windows

2. Install Python packages:

   pip install fastapi uvicorn python-multipart pillow pytesseract easyocr pdf2image python-docx PyPDF2 opencv-python numpy torch torchvision aiofiles

3. Create directory structure:

   mkdir -p models/easyocr models/tesseract temp uploads cache

Usage Guide

Starting the OCR Server

Launch the OCR processing server with:

python ocr_api.py

The server will start on http://localhost:8000 and provide both the API endpoints and web interface.

Using the Web Interface

1. Open your web browser and navigate to http://localhost:8000
2. Drag and drop files or click to select documents for processing
3. Click the “Extract” button to begin OCR processing
4. View results in the text panel or download as a text file

API Usage

The system provides RESTful endpoints for developers:

# Check server status
curl http://localhost:8000/health

# Process a single file
curl -X POST -F "[email protected]" http://localhost:8000/extract-text

# Process multiple files
curl -X POST -F "[email protected]" -F "[email protected]" http://localhost:8000/extract-text-batch

Licensing Information

This project is licensed under the GNU General Public License v3.0. This means:

• You are free to use, modify, and distribute this software
• You must disclose your source code when distributing modified versions
• You must include the original copyright notice and license terms
• Any modifications must be released under the same license
• This software comes with no warranty

The full license text is included in the LICENSE file in the project repository.

Support

For issues and questions:

1. Check the troubleshooting section in the documentation
2. Examine the application logs for error details
3. Ensure all system dependencies are properly installed
4. Verify sufficient disk space is available for model storage

Contributing

Contributions are welcome! Please feel free to submit pull requests, report bugs, or suggest new features through the project’s issue tracker.

---

Disclaimer: This software is provided as-is without any warranty. The authors are not responsible for any data loss or security issues that may arise from its use. Always maintain backups of important documents.