Multi-Stage CNN Architecture for Face Mask Detection
Amit Chavda, Jason Dsouza, Sumeet Badgujar, Ankit Damani
Abstract
Coronavirus Disease 2019 (COVID-19) broke out at the end of 2019, and it's still wreaking havoc on millions of people's lives and businesses in 2020. There is an upsurge of uneasiness among people who plan to return to their daily activities in person, as the world recovers from the pandemic and plans to get back to a state of regularity. Wearing a face mask significantly reduces the risk of viral transmission and provides a sense of protection, according to several studies. However, manually tracking the implementation of this policy is not possible. The key here is technology. We present a Convolutional Neural Network (CNN) based architecture for detecting instances of improper use of face masks. Our system uses two-stage CNN architecture that can detect both masked and unmasked faces and is compatible with CCTV cameras. This will aid in the tracking of safety violations, the promotion of face mask use, and the creation of a safe working environment.