Deep learning-based recognition of wearing masks in public during the pandemic

Abstract

Human face detection is a computer vision application. Face image processing has been the subject of various studies. Several researchers have previously investigated facial recognition. We used IOT and AI algorithms with the basic notion of human face identification in this research to identify the covid-19 patient travelling in public locations during isolation period. -19 criteria for Human face discovery is the novel notion in this covid. An Internet of Things (IoT) method is used to store daily averages of 19 positive cases across districts. The information that can be stored, such as a person's name, phone number, and address (with different poses). Personal information is securely saved in the cloud database and can be accessed at any time by logging into your account. IoT and Raspberry Pi are used to store and retrieve data. Face detection technology in CCTV cameras is used to keep tabs on the current scenario and identify any people who might be in the video. We installed cameras in strategic locations and linked them to the cloud server so that the faces of those with and those without covid 19 could be forwarded. hange detection methodologies in remotely sensed images suffer from the problem of data inadequacy; and to handle this problem, semi-supervised approaches can be opted. Semi-supervised Modified Self-organizing Feature Map is used to classify covid positive and normal cases in this recognition method. Every time a person's face is taken by the camera and compared to a database, an AI algorithm is used to identify and categorise the person (testing centre data). Covid positive patients will be flagged by an AI system, and their personal data will be sent to a government health care unit, which may take legal action against them, in this classification process. OpenCV and the Python platform were used to carry out this experiment. Public exposure to covid 19 will be reduced, and mortality rates owing to covid illness will be reduced as a result of this proposed model.