User Authentication by Face Thermograms Based on Hybrid Neural Networks
Жумажанова Самал Сагидулловна, I. I. Tatarinov · 2021
For the security of large information systems, biometric subject authentication systems are widely used. To ensure the fault tolerance of the biometric system, the most promising model of the “biometrics-to-code” converter is based on artificial neural networks. At the same time, it is important to investigate the influence of the psychophysiological state (PPS) of users on the results of their authentication. In this paper, we consider a hybrid model of a neural network “biometrics-to-code” converter (based on a new type of hybrid neural networks) that does not compromise the biometric template and the user's key (password) and is resistant to attacks of knowledge extraction, developed for biometric authentication of using thermographic images of users’ face and neck in 7 PPS.