Using CNN and YOLO Detect Face Mask and Social Distance Along With Temperature

Abstract

The COVID-19 is an unparalleled crisis leading to huge number of problems. To reduce the spread of corona virus people are advised to wear masks when surrounded by people to protect themselves. This makes face recognition very difficult since certain parts of face are hidden. Many new algorithms are devised using convolutional architecture to make face recognition accurate as possible. This convolutional architecture has made it possible to extract even pixel details.

 Designed a binary face classifier which can detect any face present in the frame irrespective of its alignment beginning from RGB image of any size.The method involves training through fully convolutional networks it detects multiple facial images in a single frame and also proposes a model to detect social distance using visualisation deep learning network. CNN and YOLO algorithms are used to detect face mask and social distance between persons respectively. Arduino controller and LM35 sensor is used to detect body temperature which alerts if temperature is above predefined value.