FACE MASK DETECTION USING CNN ALGORITHM USING DESIGN THINKING APPROACH

COVID-19 pandemic has disrupted global trade and movement, having an immediate impact on our day-to-day lives. Soon, many providers of public services will require customers to correctly wear masks to use their services. Wearing a mask lowers an infected person’s risk, regardless of whether they show symptoms. As a consequence of this, assisting society everywhere by detecting face masks has evolved into an essential task. Wearing a mask is one non-pharmaceutical method that can be used to reduce the primary source of COVID droplets released by an infected person. The goal of this paper is to create a highly accurate, real-time method for identifying faces in public where people don’t wear masks and forcing them to do so to improve community health. In this paper, we propose a strategy for recognizing individuals wearing and not wearing veils to stop the spread of Coronavirus, where all open areas are looked after by CCTV cameras. A deep learning architecture is trained with images of people wearing and not wearing masks from a variety of sources. Image processing analysis and machine learning methods can be used to assess the wear and tear on a face mask. The fundamental machine learning tools OpenCV and Scikit-Learn are utilized in this paper to simplify a strategy for achieving this objective. The face in the image is correctly identified and its mask coverage is determined using the proposed method. In addition, the risk percentage of those who are concerned will be displayed regardless of whether a mask is detected— whether it is worn correctly or not. As a performer of a surveillance task, it can also detect a face and a moving mask. The accuracy of the method can reach up to 95.77 percent.