ABSTRACT
Facial recognition is widely used for identification of people as one of the biometric authentications. Biometric authentication consists of two types physiological and behavioral features. In physiological biometrics, faces, iris, and fingerprints are used for identifying the person. In behavioral biometrics, their characteristic features namely voice, DNA and hand writing is used. While using facial recognition, an individual can be identified using the previously trained model using deep learning based on the Haar cascade algorithm. Biometric authentication has been generally used for surveillance purposes. However, due to the COVID 19 pandemic, people of each nation are in need to wear face masks for their safety. Our project uses deep learning and open cv to recognize the person and to identify whether he wears a face mask or not by using transfer learning techniques and convolution neural network. One large dataset of people with mask and people without a mask was used as a training model. Our project was able to achieve an accuracy of 96.8% during the training and testing phase. © 2022 IEEE.