Design of Distance Education Authentication System Based on Deep Convolutional Neural Netework

摘要

With the large-scale development of distance education and remote examination, the research on remote login authentication system has received more and more attention, but the traditional face recognition, such as face recognition based on geometric features, recognizes the effect when the head deflection angle or face occlusion area is large. Unsatisfactory, not suitable for the environment where the user's head and posture changes and the face is decorated. Therefore, this paper designs a face recognition system based on deep convolutional neural network and builds a real-time face recognition platform. The face detection part adopts a cascade structure of lab classifiers adapted to different head poses. Feature localization adopts a strategy of stepwise optimization of face alignment results following face image resolution. The feature extractor is composed of convolutional neural network model. Alexnet improved. The experimental results show that the remote authentication system designed in this paper can effectively identify faces that are partially obscured by various angles and faces.