Face Mask Attendance System Based on Image Recognition

EasyChair Preprint 5541

Abstract

The Pandemic of 2020 have resulted in a new normal lifestyle of wearing mask in public places and follow social distancing. Hence to develop a model for recognizing a person wearing a mask had turned mandatory. The technology, Deep Convolutional Neural Network (CNN) is used as an integral part to train a model and also to classify the images. Identifying the name of the person with the mask on is a tough process, since most of the facial characteristics are hidden by the mask. The data set of person with mask and without mask is collected and a model is trained using CNN algorithm. Real time face recognition module OPEN CV is used to capture and classify the real time facial images from the input camera to perform CNN and to compare the image with the trained model to recognize the name of the person. Image processing is widely used in many applications like image recognition, feature extraction, pattern recognition etc. Here, pre-processing is done using histogram equalization. The data set of 10 pictures per person is used for training the CNN model. The region of the eyes and position of the mask is feature extracted and it is trained to recognize the person.XML python library helps to extract the required information from the image for further processing.

Keyphrases: CNN, Open CV, Python, face mask recognizer

@booklet{EasyChair:5541,
  author    = {S Vignesh and A Sriram and G Venkatesan and A Usha},
  title     = {Face Mask Attendance System Based on Image Recognition},
  howpublished = {EasyChair Preprint 5541},
  year      = {EasyChair, 2021}}