Methods of Biometric Authentication for Person Identification

Biometric authentication methods have revolutionized the way individuals are identified and granted access to secured systems and services. Unlike traditional password-based approaches, biometrics rely on unique physical or behavioral traits to verify a person’s identity. This paper provides an overview of various biometric authentication methods used for personal identification, highlighting their characteristics and applications. The security industry places increased demands on the methods and algorithms used in the process of person identification. A spoofing attack on the user interface involves replacing a real fingerprint or face with a fake biometric image. Spoofing attacks violate the basic principle of operation of recognition systems, and system security is seriously compromised. A template database leak occurs when information about a user’s legitimate template becomes available to an attacker. In this case, it is much easier for the attacker to recover the biometric template by reverse engineering of the template, which increases the risk of forgery. However, the attacker is not able to replace the real template with a fake one, as in the case of a password, and this is an advantage of biometric methods. The article presents the results of studies of biometric authentication methods used in access control systems in terms of reliability, accuracy, and training time. The expediency of using convolutional neural networks, support vector methods, k-nearest neighbors, and the Frobenius norm is considered. Comparative characteristics are obtained, and the results of testing of the selected methods are presented.