ABSTRACT
An appropriate mask protects individuals from infectious illness and greatly minimises the spread of COVID-2019 in public spaces like institutions and temples. This needs surveillance technology capable of detecting persons wearing correctly fitted masks. However, this is not the purpose of the face detection algorithms that are currently in use.The researchers suggest a two-stage technique for identifying mask wear using hybrid machine learning algorithms in this paper. The first step involves identifying as many possible candidate locations for wearing masks as possible employing Faster RCNN and ResNet V2 structures.In comparison, the second step entails employing a massive learning system to validate the real face masks. It is achieved by the training of a model with two classes. Additionally, this article describes a data collection conducted during the Market, Malls and contains 2804 realistic images. The suggested method exceeds all other techniques that are already in use, with an accuracy rate of 99.2 percent for straightforward circumstances. Copyright © 2022, Anka Publishers. All rights reserved.