Face Sense - Real-time Computer Vision Face Recognition Attendance System

This project is a Face Recognition Attendance System developed using Python. It captures and processes real-time video from a webcam to recognize faces, logs attendance in both CSV and SQLite formats, and serves attendance data via a Flask web application.

Project Structure

Face-Sense
│
├── attendance_data/
│   ├── attendance.db             # SQLite database for attendance records
│   ├── YYYY-MM-DD.csv            # Daily attendance CSV files
│
├── model_data/
│   ├── known_names.bin           # Pickle file containing the names of known faces
│   ├── known_encodings.bin       # Pickle file containing the encodings of known faces
│
├── student_photos/
│   ├── [student_name].jpg        # Photos of students used for face encoding
│
├── templates/
│   ├── index.html                # HTML template for the Flask web application
│
├── requirements/
│   └── requirements.txt          # List of required Python packages
│
├── engine.py                     # Main script to start the Flask server and process video
├── info_storing.py               # Script to encode student faces and store encodings
└── main.py                       # Main script for face recognition and attendance logging

Installation

1. Clone the repository:

   git clone https://github.com/ishtiuk/Face-Sense.git
   cd Face-Sense

2. Set up the environment:

   Ensure you have Python 3.7 or higher installed. Install the required Python packages using the requirements.txt file:

   pip install -r requirements/requirements.txt

3. Prepare the model data:

   Ensure you have the student photos in the student_photos/ directory. Run the info_storing.py script to generate face encodings and save them:

   python info_storing.py

Usage

Running the Application

1. Start the Flask server:

   Run the engine.py script to start the Flask server and initiate the face recognition process:

   python engine.py

   The server will be accessible at http://localhost:5000.

2. View the video feed and attendance data:

   • Open your web browser and navigate to http://localhost:5000 to view the live video feed from the webcam.
   • The video feed will display the recognized faces with their names and confidence scores.
   • The attendance data can be accessed via the /get_attendance_data endpoint.

Stopping the Application

To stop the application, press q in the terminal where the video feed is displayed or terminate the Flask server using Ctrl+C.

Code Explanation

• engine.py: Contains the Flask application and the main face recognition logic. It captures video from the webcam, processes each frame to detect and recognize faces, and logs attendance data to both a CSV file and an SQLite database.

• info_storing.py: Prepares the model by encoding student faces and saving the encodings and names to binary files. Run this script once to prepare the face recognition model.

• main.py: (Not actively used in this setup but similar to engine.py for standalone execution without Flask)

Notes

• Ensure the webcam is properly connected and accessible by the application.
• Modify the known_names and known_encodings files if you add or remove students from the student_photos/ directory.
• The system assumes that student names in the student_photos/ directory are formatted as [first_name]_[last_name].jpg.

Troubleshooting

• Issue: Faces not being recognized or attendance not being logged.

  • Solution: Check the student photos and ensure they are clear and correctly formatted. Re-run info_storing.py to update the encodings.

• Issue: Flask server fails to start.

  • Solution: Ensure all required packages are installed and that no other process is using port 5000.

License

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