Open AccessFace Mask Detection with Raspberry Pi
Author(s) -
G. Pavan Kumar
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.35778
Subject(s) - raspberry pi , computer science , artificial intelligence , computer vision , covid-19 , facial recognition system , face (sociological concept) , deep learning , frame (networking) , pixel , medicine , pattern recognition (psychology) , computer security , infectious disease (medical specialty) , telecommunications , social science , disease , pathology , sociology , internet of things
In the wake of the COVID-19 epidemic, institutions such as the academy are suffering the most from global closure if the current situation haven’t rectified. COVID-19 also known as Serious Acute Respiratory Syndrome Corona virus-2 is an infectious disease that is transmitted to an infected person who talks, sneezes or coughs through respiratory droplets. This spreads quickly through close contact with anyone with the disease, or by touching objects or the infected area. By wearing a face mask under the jaws covering at crowded places or by frequently hygiene at your palms and by using at the minimum of 70% sanitizers which are based on alcohol is the best method for the against of the COVID-19. In this project we have used it ML, OpenCV and TensorFlow face recognition. This the model can be used for security purposes because of course an app that works well for use. In this way MobilenetV2 using a BN-based layout too lightweight and embedded this model with Raspberry pi to make real-time mask discovery, when, SSD (Single Shot Detector) format is used and the spinal network is light. As technology advances, Deep Learning has demonstrated its effectiveness in recognition and classification through image processing. The study uses in-depth reading techniques to distinguish facial recognition and to determine whether a person is wearing a facemask or not. The collected data contains 25,000 images using 224x224 pixel resolution and obtained 96% accuracy with the performance of a trained model. The system enhances the Raspberry Pi-based real-time recognition made by alarms and takes a facial image when the person found is not wearing a facemask. This study is beneficial in combating the spread of the virus and in avoiding contact with it.