Online attendance system based on facial recognition with face mask detection

Area of Science:

• Computer Science
• Biometrics
• Artificial Intelligence

Background:

• Traditional attendance systems are often manual and time-consuming.
• Biometric systems offer enhanced security and efficiency.
• The COVID-19 pandemic highlighted the need for contactless solutions, including mask detection.

Purpose of the Study:

• To develop an online attendance system leveraging facial recognition and facial mask detection.
• To provide a web-based interface for universal accessibility without software installation.
• To establish a centralized online database for efficient attendance data management.

Main Methods:

• Utilized Support Vector Machine (SVM) for facial recognition model training.
• Incorporated synthetic data for training the model to detect faces with masks.
• Developed the server application in Python with OpenCV for image processing.
• Employed PHP and MySQL for web interfaces and database management.

Main Results:

• Achieved approximately 81.8% accuracy for facial recognition using a pre-trained model.
• Attained around 80% accuracy for facial mask detection.
• Demonstrated successful integration of Python and PHP for online server processing.

Conclusions:

• The developed system offers an effective and accessible online attendance solution.
• The integration of facial recognition and mask detection enhances attendance tracking capabilities.
• The browser-based interface ensures ease of use across various terminals.