APPLICATION OF DEEP LEARNING: FACE RECOGNITION FOR VERIFICATION OF STUDENT IDENTITY IN THE EXAM ROOM

Abstract

Face recognition is one of the critical areas of computer vision, which aims to verify a person's identity based on images or videos. Face recognition is applied in many fields such as security systems, biometric systems, attendance, etc. Many face recognition techniques have been researched and developed, in which deep learning techniques give outstanding accuracy. This paper proposes a model based on Convolutional Neural Network (CNN) to recognize faces from images to verify student identity when entering the exam room. First, we use the MTCNN algorithm for face detection and data preprocessing. Then, the results will be fed into the FaceNet model, a Google model based on CNN, for feature extraction and use the Triplet loss function to optimize the recognition. The student image dataset (STUDUE) is built for this study. Experiments were performed on the Yale and STUDUE image dataset with the accuracy of 92.1% and 88.4%, respectively. The experimental results are compared with other studies on the same image dataset, showing the accuracy and efficiency of the proposed model