🔍 Frappe OCR – AI‑Powered Document Scanner

Transform your document scanning workflow with AI‑powered OCR technology.

Features • Installation • Usage • API • Contributing

---

🚀 What is Frappe OCR?

Frappe OCR is a passport‑scanning and text‑extraction app for the Frappe Framework. It combines traditional OCR (Tesseract) with Google’s Gemini AI for accurate, intelligent document processing.

🎯 Ideal for:

• Government Services – Passport applications and renewals
• Travel Agencies – Fast customer data entry from travel documents
• Banks & Financial Institutions – KYC document processing
• Immigration Services – Automated document verification
• Any Business – General document digitization

---

✨ Features

🧠 AI‑Powered Intelligence

• Gemini AI Integration – Intelligent parsing and error correction
• Smart Error Correction – Automatically fixes common OCR mistakes
• Contextual Understanding – Extracts structured data (e.g., MRZ fields)

📄 Document Support

• Multi‑format – PNG, JPG, JPEG, PDF
• Passport‑Optimized – MRZ (Machine Readable Zone) parsing
• Batch Processing – Scan multiple files at once

🔧 Advanced OCR Pipeline

• Image Preprocessing – Enhancement and noise reduction
• Adaptive Thresholding – Robust to varying lighting
• Skew Correction – Auto‑straighten tilted scans

💾 Frappe Integration

• Native DocTypes – Leverage File Manager with full Frappe features
• Role‑Based Access – Fine‑grained permissions
• Search & Filter – Full‑text search on extracted content
• RESTful API – Integrate with external systems

---

🛠 Installation

Prerequisites

Install Tesseract and development headers:

# Ubuntu/Debian
sudo apt update && sudo apt install tesseract-ocr libtesseract-dev

# macOS (Homebrew)
brew install tesseract

# CentOS/RHEL
sudo yum install tesseract tesseract-devel

Install the App

# In your bench directory
cd /path/to/bench

# Clone and install
bench get-app https://github.com/DeliveryDevs-ERP/ocr.git
bench --site your-site install-app ocr

# Install Python dependencies
bench setup requirements

# Restart bench
bench restart

Configuration

Create a .env file at the app root:

# Gemini AI settings
GEMINI_API_KEY=your_gemini_api_key_here
GEMINI_MODEL=gemini-2.0-flash

🔑 Get your Gemini API Key: Visit Google AI Studio.

---

📖 Usage

🖥 Web Interface

1. Open the OCR module in your Frappe Desk.
2. Click New to create a File Manager record.
3. Upload an image or PDF.
4. Click Save – OCR runs automatically.
5. Review extracted text in Scanned Contents.

📱 API Usage

import frappe

# Scan a passport image
parsed, raw = frappe.call(
    'ocr.ocr.doctype.file_manager.file_manager.scan_passport',
    file_url='/files/passport_scan.jpg'
)
print(parsed)
print(raw)

Sample Response

{
  "passport_type": "P",
  "last_name": "SMITH",
  "first_name": "JOHN",
  "passport_number": "AB1234567",
  "nationality": "USA",
  "date_of_birth": "1985-06-15",
  "sex": "M",
  "date_of_expiry": "2030-12-20",
  "cnic": "1234567890123"
}

---

🎮 Demo

Before OCR ➡️ After OCR

Raw Scan	Extracted Data
	json{"first_name": "JOHN", "last_name": "SMITH", "passport_number": "AB1234567"}

---

🔌 API Reference

scan_passport(file_url)

Extracts structured data from passport files.

Parameters:

• file_url (string): Path to the uploaded file

Returns:

• parsed_data (dict): Structured fields
• raw_text (string): Full OCR output

---

🤝 Contributing

1. Fork this repo
2. Create a feature branch (git checkout -b feature/XYZ)
3. Commit your changes (git commit -m "feat: add XYZ")
4. Push to branch (git push origin feature/XYZ)
5. Open a Pull Request

We welcome contributions of all kinds! Feel free to open issues or submit PRs.

---

---

© 2025 Deliverydevs. Licensed under MIT.