Mask Compliance in Public Services: A Comparative Study of Machine Learning and CNN Models for Mask Detection

The Coronavirus Disease (COVID)-19 pandemic is fast changing our way of life and interfering with international travel and trade. The norm now is to wear a protective facemask. Many public service providers may soon demand users to correctly wear masks in order to access their services. This project offers a condensed method for accomplishing this goal by utilizing CNN models like VGG19, SVM, and Decision Trees as well as machine learning models. With an accuracy of 97%, the first methodology used Resnet50 as a feature extractor and the SVM + Decision tree (combined) - Stacking ensemble method as a classifier. Using the acquired dataset, the second methodology used the VGG19 model, which produced an impressive accuracy of 94.7%. The study will also look into how other elements, such lighting, camera angles, and facial orientations, affect how well the algorithms work. The results of this study can be applied to the development of more accurate and dependable mask detection systems, thereby reducing the transmission of infectious diseases in public areas.