Face recognition algorithm and edge computing technology based on deep learning

With the rapid development of deep learning, more and more attention has been paid to the research and application of face recognition and expression recognition in this era of big data. In security monitoring, human-computer interaction and other aspects, the related applications are more and more. At present, the common face recognition system mainly has two ways of work, offline work and network work. Offline face recognition system algorithm calculation speed is fast, but the recognition accuracy is low; The algorithm recognition accuracy of the face recognition system working in the network is high, but the calculation and transmission time is high. Aiming at the above problems, this paper focuses on the design of two intelligent human face recognition systems. Firstly, a face verification system with fast computation speed and high recognition accuracy is designed and developed on a small embedded board by using deep face recognition neural network and model compression technology. Secondly, using face recognition algorithm, edge computing technology and enhancement learning Leavening algorithm, design and develop an intelligent video surveillance system that works in the network. The system can process the monitoring data intelligently through the synergistic effect of the front and rear end face recognition algorithm and the decision algorithm of the decision-maker module, so as to solve the problem that the traditional monitoring system occupies too much storage and computing resources and can not respond in time. Finally, this paper introduces some work on facial expression recognition. An expression recognition method based on the fusion of multiple local features is proposed. The algorithm improves the accuracy of facial expression recognition and solves the problem of poor performance of traditional single expression feature recognition.