Application of Artificial Intelligence in Monitoring the Use of Protective Masks

Abstract

In the context of current epidemic diseases, this study developed a web application, which can monitor the use of protective masks in public environments. Using the Flask framework in Python language, the application has a control panel to help visualize the obtained data. In the detection process, Haar Cascade algorithm is used to classify faces with and without protective masks. Therefore, the web applications are lightweight, allowing the detection and storage of images captured in the cloud and thte possibility of further data analysis. The classifier presents precision, reversal and f-score of 63%, 93% and 75%, respectively. Although the accuracy is satisfactory, new experiments will be carried out to explore new computer vision technologies, such as the use of deep learning.