edgeidx 1.0.4

EdgeIDX is a Windows-only Python library for OCR that extracts information from Aadhaar and PAN cards with embedded Tesseract engine.

edgeIDX

Overview

edgeIDX is a Python package for extracting Aadhaar card details (Name, DOB, Gender, and Aadhaar Number) from images using Tesseract OCR. Unlike traditional setups, this package embeds a trained Tesseract model within itself, eliminating the need for a separate `tesseract.exe` installation.

This project aims to address the limitations of existing OCR systems by focusing on accuracy, speed, and lightweight deployment. It is specifically designed for extracting information from government-issued identity documents such as Aadhaar and PAN cards.

Features

• Extracts Aadhaar card details (Name, DOB, Gender, and Aadhaar Number) from images using OCR.

• Uses an embedded Tesseract model, requiring no external dependencies.

• Supports image preprocessing to improve OCR accuracy.

• Works across different platforms without requiring external Tesseract installation.

• Can be integrated with Android applications for automated Aadhaar data extraction.

• High Accuracy OCR: Achieves over 95% accuracy in extracting key details such as names, dates of birth, and ID numbers.

• Real-Time Processing: Processes each document within 1-2 seconds to facilitate high-throughput applications.

• Edge Device Compatibility: Designed for low-resource environments like mobile phones.

• Open-Source and Developer-Friendly: Provides a modular, open-source library with clear documentation.

• JSON Output Format: Extracted details are formatted in JSON for seamless integration into different applications.

• Compliance with Data Privacy Regulations: Ensures that the extracted information is handled securely and adheres to laws such as Aadhaar Act, 2016, Personal Data Protection Bill, 2021, and General Data Protection Regulation (GDPR).

• Future-Proof Design: The modular architecture allows easy integration of additional document types in future releases.

System Requirements

• Python 3.7+
• OpenCV for image processing
• NumPy for handling image arrays
• Pytesseract for OCR
• Supported Operating Systems: Windows and Linux.
• Supported Edge Devices: Mobile phones, Raspberry Pi, and other low-power devices.
• Development Frameworks: OpenCV, and Tesseract-OCR.

Installation

1. Install via pip

```sh
pip install edgeidx
```

Usage

```python
#   Example Usage (Adapt this to your library's API)
from edgeidx.ocr import extract_aadhaar_details

image_path = "sample_13.jpg"
print(extract_aadhaar_details(image_path))
```

Output Format

The extracted data is returned in JSON format:

```json
{
    "name": "John Doe",
    "date_of_birth": "01/01/1990",
    "gender": "Male",
    "aadhaar_number": "XXXX-XXXX-XXXX"
}
```

Constraints

• Compliance with Data Protection Laws: The system must follow Aadhaar Act, GDPR, and data privacy regulations.
• Lightweight Execution: The model must be optimized for low-power devices through pruning, quantization, and efficient memory usage.

Non-Functional Requirements

• Performance:

  • Ensure OCR accuracy > 95%.
  • Processing time per document: 1-2 seconds.

• Usability:

  • Provides clear API documentation and sample integration examples.

• Software Quality Attributes:

  • Efficiency: Optimized for low-latency execution.
  • Scalability: Supports additional document types with minimal modifications.

Use Cases

• Developers: Integrate edgeIDX into their applications using provided APIs and libraries.
• Businesses: Automate document verification processes to improve efficiency in sectors like banking, healthcare, and telecom.
• Prototype Developers: Using Python Library for research

Team

• Nawaz Bashir Sayyad
• Kunal Chandak
• Mayuresh Muluk
• Harshal Pathare

References

• Aadhaar Act, 2016: Legal guidelines for handling Aadhaar data in India.
• Personal Data Protection Bill, 2021: Regulations related to sensitive personal data processing.
• General Data Protection Regulation (GDPR): European data privacy and protection regulations.

Conclusion

edgeIDX is an open-source, high-accuracy OCR system optimized for government-issued documents. It offers fast processing, modular design, and edge-device compatibility, making it a valuable contribution to the open-source community.