Hazzer — Face Recognition Attendance System

A Flask-based employee attendance tracking system that uses facial recognition (DeepFace / ArcFace) to log staff check-in and check-out times. Employees are registered once with a photo, and afterward the system recognizes their face to automatically record their attendance.

Features

• Register new employees with a profile photo
• Face recognition for check-in and check-out
• Attendance reporting
• Simple, lightweight web interface

Requirements

• Python 3.9 or higher
• A webcam for face scanning

Installation

Clone the repository:

git clone https://github.com/GafelSon/hazzer-py.git
cd hazzer-py

Install the dependencies:

pip install tensorflow==2.21.0 tf-keras
pip install -r requirements.txt

Install TensorFlow and tf-keras first, before the rest of the dependencies, since DeepFace relies on them.

Running the App

python app.py

Once running, the app is available at:

http://localhost:5005

Usage

1. Go to the employee management section and add a new employee along with a clear photo of their face.
2. Use the scan page to check in or out — the system identifies the face with DeepFace (ArcFace) and logs the attendance record.
3. View attendance reports under the Reports section.

Project Structure

hazzer-py/
├── app.py              # Main Flask app and routes
├── database.py         # Database handling for employees and attendance records
├── face_utils.py        # Face recognition utility functions using DeepFace
├── static/              # Static files (CSS, JS, images)
├── templates/           # HTML templates (Jinja2)
├── requirements.txt      # Python dependencies
└── .gitignore

Tech Stack

• Flask — backend web framework
• DeepFace (ArcFace) — face recognition and matching
• OpenCV — image processing and webcam handling
• TensorFlow / tf-keras — deep learning model backend

Notes

• Registration photos should be clear and well-lit for the best recognition accuracy.
• The project currently ships with a Persian (Farsi) localized dashboard.

Contributing

Found a bug or have a suggestion? Feel free to open an Issue or submit a Pull Request.

License

This project is licensed under the MIT License — see the LICENSE file for details.