Covid-19 started in November 2019 and started to spread across the world killing 3.74 million people. To curb the spread were to wear masks and maintain 6 feet distance. Although vaccines are available now and the cases have come down in some few countries, many countries are still struggling. In order to aid in stopping the spread and identify individuals not following the safety policies, we aim to build an Object Detection and Convolution Neural Network based face mask and social distance detection system. The dataset to be used contains numerous images of instances where people are with and without mask and the model aims to identify people violating safety policies and flag the images with violation concerns.
MODEL 1 : CONVOLUTIONAL NEURAL NETWORK
MODEL 2 : CNN with DATA AUGMENTATION
MODEL 3: MOBILENETV2
DATA SET : Web scraped dataset using selinium containing mask and no mask images
