Face Detection-Induced Access Control System via Large Margin Metric Learning

Abstract

With the development of science and technology and the acceleration of economic integration, identity authentication has become the most basic element in cyberspace and the basis of the whole information security system. Biometric recognition technology is an important technology in the process of identity authentication. Among them, face recognition technology has been favored by researchers, social applications, and users in the field of identity authentication by virtue of its inherent advantages such as ease of use and insensitivity. In this paper, a face recognition-based access control system is established with the help of large margin metric learning. First, a face library is input into a deep neural network to extract representation features. Second, the deep representation features are used to learn a large margin metric learning model. Third, the face image is captured by a digital camera to input into large margin metric learning model for identifying the person. The experimental results show that the proposed system can accurately identify most of the persons.