International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 06 | June 2022 www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
FACE MASK DETECTION USING ML Prof .V.O.Patil, Darshana Rajendra Javheri, Neha Vinod Patil, Aditi Sanjay Jaiswal 1Assistant Professor at Department of Computer Engineering, PSGVP Mandal’s D.N.Patel College of Engineering, Shahada, Maharashtra, India 2Final Year Students of Department of Computer Engineering, PSGVP Mandal’s D.N.Patel College of Engineering, Shahada, Maharashtra, India ------------------------------------------------------------------***--------------------------------------------------------------------------Abstract- As we all know COVID-19 pandemic has rapidly affected our day-to-day life disrupting the world trade and movements.
So wearing a mask, wearing a protective face mask is become the new normal. However, still have few ways to control the outbreak as instructed by the WHO (World Health Organization).Hence Using Face Mask Detection System, one can monitor if the people are wearing masks or not. This project presents a simplified approach to achieve this purpose using some basic Machine Learning packages like TensorFlow, Keras, Open, CV and Scikit-Learn. There are a few processes involved in achieving the objective of the project that includes pre-processing, data augmentationn, training ,testing, and image segmentation. The method attains accuracy up to 95.77 percent and 94.58 percent respectively on two different datasets.
Keywords- Tenserflow, openCV, Keras, Numpy, Scikit-Learn. I. INTRODUCTION COVID-19 pandemic has had an enduring effect in many nations global due to the fact that December 2019. It originated in Wuhan, China. The World Health Organization (WHO) as of March 11, 2020, declared it a pandemic that received its roots throughout the globe and critically affected 114 countries. Every clinical professional, healthcare organizations, clinical practitioners, and researchers are searching for right vaccines and drugs to conquer this lethal disease, but no leap forward has been stated until date. The virus spreads thru air channel whilst an inflamed man or woman sneezes or talk with the alternative man or woman, the water droplets from their nostril or mouth disseminate thru the air and have an effect on different peoples withinside the vicinity (Kumar et al., 2020). Face Mask detection has turn out to be a trending software because of the Covid-19 pandemic, which needs someone to put on face masks, maintain social distancing, and use hand sanitizers to scrub their hands. While different issues of social distancing and sanitization were addressed till now, the problem of face masks detection has now no longer but been accurately addressed. Wearing a masks all through this pandemic is a essential safety measure and is maximum crucial step in instances whilst social distancing is tough to maintain. Face Mask detection has became as much as be an outstanding hassle withinside the area of photograph processing and pc vision. Face detection has diverse use instances starting from face popularity to shooting facial motions, in which the latter requires the face to be discovered with very excessive precision. Due to the speedy development withinside the area of device gaining knowledge of algorithms, the jeopardies of face masks detection era appear to be properly addressed yet. This era is extra applicable these days because it is used to detect faces not only in static images and videos but also in real-time inspection and supervision.
II.
LITURATURE SURVEY
In the latest past, diverse researchers and analysts specifically targeted on gray-scale face image. While a few had been absolutely constructed on sample identity models, owning preliminary facts of the face version whilst others had been the use of AdaBoost, which changed into an fantastic classifier for schooling purposes. Then got here the Viola-Jones Detector, which supplied a leap forward in face detection technology, and real-time face detection were given possible. It confronted diverse issues just like the orientation and brightness of the face, making it tough to intercept. So basically, it did not paintings in stupid and dim light. Thus, researchers began out looking for a brand new opportunity version that might effortlessly discover faces in addition to mask at the face.
© 2022, IRJET
|
Impact Factor value: 7.529
|
ISO 9001:2008 Certified Journal
|
Page 2942