Maskify: Precision Detection and Fit Assessment for Mask Compliance

We introduce Maskify, an innovative tool designed to automatically detect and evaluate the use of face masks. Amid the challenges of the COVID-19 pandemic, wearing masks has emerged as a crucial strategy to limit the virus’s spread, especially in environments where mask mandates are enforced. Reliable systems for monitoring mask usage are therefore vital for public safety.Our solution leverages deep learning to not only determine if a person is wearing a mask but also to assess how well the mask fits using a specialized metric called the "wearable score." This score quantifies the extent to which the mask properly covers essential areas of the face. Developing Maskify involved creating a diverse dataset that includes images of both masked and unmasked individuals.The process begins with advanced pre-trained models that accurately identify faces in images. Once a face is detected, key landmarks on the lower face are extracted to assist in classifying the mask. We utilize ResNet-based architectures, known for their robust image classification performance, to enhance the detection accuracy. The wearable score is then computed by segmenting the mask region and calculating the percentage of the face that is effectively covered.In summary, Maskify offers a robust approach for monitoring mask compliance and ensuring proper fit, thereby supporting ongoing public health efforts to mitigate the spread of COVID-19.