Masked & Unmasked Face Recognition Using Support Vector Machine Classifier

Abstract

Face masks become a need in epidemic scenarios such as the Corona virus pandemic of 2020-21. Most companies prefer face authentication instead of fingerprint, signature, and card verification. Face mask gives protection against Corona virus than other traditional methods used for identification. In the case of facial recognition, a machine must detect and recognise the face in a picture. In this paper used methods are supported by machine learning that permits a machine to evolve through a learning process and to perform recognition tasks. Caffe model of deep learning is used for face detection. The training dataset contains both masked and non-masked faces. This project and outcome has developed an approach to recognize faces in a real-time video stream that can also be used in the existing recognition systems to identify masked faces. Facial recognition has been done with a Support Vector Machine classifier. All are implemented in Python with OpenCv with tools modules and sub-modules.