Face mask recognition system using CNN model.

Area of Science:

• Computer Science
• Artificial Intelligence
• Public Health

Background:

• The COVID-19 pandemic necessitated widespread face mask usage, making appropriate mask detection essential for public safety and service access.
• The shift to mask-wearing as a norm requires reliable automated systems for monitoring compliance in various settings.

Purpose of the Study:

• To develop and evaluate a machine learning-based system for accurate face mask detection in images and videos.
• To investigate the optimization of Convolutional Neural Network (CNN) parameters for precise mask identification while preventing overfitting.

Main Methods:

• Utilized fundamental machine learning tools including TensorFlow, Keras, OpenCV, and Scikit-Learn.
• Developed a technique to detect faces in images/videos and subsequently classify mask presence.
• Employed Convolutional Neural Networks (CNNs) and explored parameter tuning for optimal performance.

Main Results:

• The proposed technique successfully recognizes faces and determines mask status in static images and dynamic videos.
• The system demonstrated excellent accuracy in identifying individuals wearing masks.
• Optimal CNN parameters were identified to enhance detection accuracy and mitigate overfitting.

Conclusions:

• The developed machine learning approach offers a simple yet effective solution for automated face mask detection.
• This technology is vital for public health surveillance and ensuring compliance with mask mandates.
• The study highlights the potential of CNNs in accurately identifying mask usage in real-world scenarios.