Deep Learning Based Facial Mask Detection Using Mobilenetv2

The Covid-19 pandemic has had a profound effect on our daily lives. One of the most effective ways to protect ourselves from this virus is to wear face masks. This research paper introduces face mask detection that authorities can use to reduce and prevent COVID-19. The face mask recognition process in this research paper is done with a deep learning algorithm and image processing done using MobileNetV2. Steps to build the model are data collection, pre-processing, data classification, model training and model testing. The authors came up with this approach due to the recent Covid-19 situations for following specific guidelines and the uprising trend of Artificial Intelligence and Machine Learning and its real-world practices. This system has been made to detect more than one person whether they are wearing masks or not. This system also gives us the Covid cases-related worldwide updates as per our chosen country and type of cases like total cases, total deaths etc. Such systems are already available, but the efficiency of the available mask detection systems was not achieved thoroughly. This newly developed system proposes to take a step further, which recognizes more than one person at a time and increases the accuracy level to a much greater extent. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Vision-Based Real-Time Hand Wash Accuracy Prediction

Hands are the primary means of germ transmission, which results in the transmission of deadly diseases. Hand hygiene is thus an important measure to prevent the spread of harmful diseases. The World Health Organization (WHO) has recommended seven steps for proper handwashing hygiene. However, not everyone adheres to the WHO’s handwashing guidelines of seven steps. A proper hand wash is an important factor in protecting people’s health during the corona virus disease (COVID-19) pandemic, especially for healthcare workers who are exposed to bacteria, influenza, and other infectious diseases. There are technologies available like ultraviolet (UV) images to check a person’s hand hygiene. However, there is no real-time system to monitor how efficiently a person makes his or her hand wash. A real-time monitoring system is needed to check people’s hand wash hygiene in public places and hospitals, reducing the risk of spreading communicable diseases. In this paper, we have discussed predicting the accuracy of handwashing actions performed by humans using a deep learning model with transfer learning (VGG16). © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.