Somya Jaiswal, Suryansh Yadav
COVID-19 is spreading all over the world, we need to stay safe by taking some simple precautions, such as physical distancing, wearing a mask, keeping rooms well ventilated, avoiding crowds, and cleaning our hands. According to World Health Organization (WHO) guidelines, when we wear a mask, we protect others as well as ourselves. Masks work best when everyone wears one. Masks should completely cover the nose and mouth and fit snugly against the sides of the face without gaps. Reports show that wearing a face mask reduces the risk of transmission[2]. Still, many people are not following these guidelines. So, we need an efficient solution for monitoring these people and creating a safe environment. Smart Brains is a face mask detection system which detects whether people are wearing a face mask or not and alerts everyone in the surrounding if someone is found without a face mask. In this project, we have used the MobileNetV2 Convolutional Neural Network (CNN) developed by Google to train a classifier model that can classify faces with masks and without masks in the live video stream. For training purposes, the dataset we have used consists of images with masks and without masks. After training, we have deployed this model on Raspberry Pi using its camera module to detect the faces without a mask on a live video stream. If any face is found without a proper mask, the system will alert people in the surrounding using a buzzer sound.
Deep Learning, Computer Vision, Tensorflow, Scikit-Learn, OpenCV, Keras, MobileNetV2, RaspberryPi