A Deep Learning Classification Algorithm for Facial Mask Detection

Abstract

In order to make the representation more meaningful, the deep learning algorithm represents data in
layers of learning layers. In deep learning, "deep" refers to the start of layers of sequential representation. This
study is to serve as a guide for developing a system and evaluating the outcomes of face mask identification
using a deep learning algorithm. Because it addresses a critical aspect of public health and safety, research on
face mask detection is extremely significant. It is essential for encouraging compliance with mask-wearing
regulations, reducing the spread of infectious diseases, and providing useful information for policy evaluation
and monitoring. Furthermore, this field of research has gained more importance and attention in the field of
public health and safety, particularly in the wake of the COVID-19 pandemic since 2020, when wearing a mask
has been widely recommended as a means of preventing the virus from spreading. According to the findings of
the study, this model is capable of accurately identifying faces, including those of people who are not wearing
masks. The average sensitivity or recall value of 93.47% and the average specificity and precision of 96.00%
make this clear. Furthermore, with an average classification accuracy of 94.73%, this model has also shown itself
to be highly accurate