QuickMark: Attendance System with Dual Authentication

Introduction

QuickMark is an attendance system designed to streamline the attendance tracking process using dual authentication methods: face recognition and QR codes. The system utilizes Tkinter for the frontend user interface and integrates with a database for efficient attendance maintenance.

Features

• Dual Authentication:

  • Face Recognition: Utilizes facial recognition technology to verify and record attendance.
  • QR Codes: Generates and scans QR codes for an additional layer of authentication.

• User-Friendly Interface:

  • Built with Tkinter to provide a simple and intuitive frontend for users.

• Database Integration:

  • Stores attendance data in a database for easy maintenance and retrieval.

• Anti-Spoofing Technology:

  • Eye Blink Detection: Ensures the person is real by detecting natural eye blinking.
  • Head Movement Detection: Verifies liveness through random head movement challenges.
  • Depth Estimation: Uses MediaPipe to detect flat surfaces like printed photos.
  • Challenge-Response System: Randomized prompts requiring specific actions to verify presence.

Requirements

Ensure you have the following dependencies installed:

pip install face_recognition opencv-python pyzbar qrcode pillow mediapipe numpy

Alternatively, install all requirements:

pip install -r requirements.txt

Setup and Usage

1. Setting up Face Data

1. Run register_face.py to capture image of student
2. Run create_face_data.py:
3. Enter the name of student exactly as it appears in their photo filename
4. The script will:
   • Create image encodings using face_recognition module
   • Store it in face_data.pkl file

2. Running the Application

1. Launch the application:

   python main.py

2. The system will:
   • Open your webcam
   • Present random liveness challenges (blink, turn head, etc.)
   • Analyze facial depth to detect spoofing attempts
   • Perform face recognition
   • Mark attendance only when all verification steps pass
3. Press 'q' to exit the application

3. Viewing Attendance

• Attendance data is automatically stored in attendance.db
• Use the UI to view attendance records

File Structure

• face_data.pkl: Stores face recognition data
• attendance.db: SQLite database for attendance records
• classroom/: Directory for student photos (in repository root)

Troubleshooting

1. If face recognition fails:

   • Ensure photos are clear and well-lit
   • Check that face is clearly visible in the photo
   • Verify photo naming matches exactly

2. If anti-spoofing checks fail:

   • Make sure there's sufficient lighting for depth detection
   • Follow the challenge instructions carefully
   • Ensure your face is clearly visible to the camera
   • Avoid quick movements that may interfere with tracking

Contributions

Contributions are welcome! If you find any issues or want to enhance the functionality, please feel free to create a pull request.

License

This project is licensed under the MIT License

---

Enjoy using QuickMark! If you have any questions or feedback, feel free to contact me!