Face Mask Detection Exploiting CNN and MobileNetV2

The whole world was suffering from the COVID-19 pandemic since 2019, which originated in the city of Wuhan, China and has quickly spread various countries, with many cases having been reported worldwide. Around 31.3crores cases had been recorded worldwide, among which India has been recorded 3.59crore cases according to recent data. Each time, this virus comes into existence with a new variant or strain. Recent studies have shown that through proper use of masks this kind of viruses can be prevented from spreading. Every nation is attempting to stop the disease's spread. Each individual must put on a mask in a public area in order to remedy the issue. Therefore, using a convolutional neural network, we propose a model that can distinguish between masked and non-masked faces. This investigation approaches to check whether a person is wearing a proper mask or not, and the mask is secure or not towards these viruses. This system mainly can be used in different transport system such as public bus, trains, aeroplane, etc., where bus conductors, ticket checker in trains and air hostage, respectively, did not had to go to near the passengers to ask them for using a proper mask. Based on CNN and MobileNetV2, a deep learning (DL) framework, this model is capable of identifying individuals without masks. It is accurate to a maximum of 99.76% using MobileNetV2 and 99.83% through CNN approach. A model has been compared with state-of-the-art methods which will be used to track the use of masks in locations including schools, offices, and other public spaces.