DOI: https://doi.org/10.36719/2663-4619/114/298-301
Nabi Huseynov
Odlar Yurdu University
Master student
https://orcid.org/0009-0004-2086-8621
huseynovnebi624@gmail.com
Improvement the Resilience of Facial Recognition Systems
Against External Attacks
Abstract
This article explores methods to improve the resilience of face recognition systems against external attacks. External attacks attempt to deceive the system by making small changes in the data to mislead the face recognition models. Face recognition technology is widely used in fields such as security, identification, and authentication; however, it is crucial to protect this technology from external attacks to ensure privacy and security. This research aims to enhance the reliability and security of the system by studying the weaknesses of existing face recognition systems and applying defense strategies against various types of external manipulations. To achieve this, the vulnerabilities of deep learning models are analyzed, and AI-based security mechanisms are proposed to make them more resilient. Additionally, adaptive defense strategies are developed to improve the system's resilience against different attack types. With these approaches, the performance of face recognition systems will be more stable and reliable, and the existing security vulnerabilities will be mitigated. The research will enable the development of stronger and more attack-resistant systems through the implementation of innovative solutions in this field.
In this context, the article presents a comprehensive study, referring to works by both local and international authors, and providing precise facts.
Keywords: facial recognition, external attacks, system reliability, modelling, security