USING THE HUMAN FACE RECOGNITION METHOD BASED ON THE MOBILENETV2 NEURAL NETWORK IN AUTHENTICATION SYSTEMS

The rapid development of biometric authentication systems has led to the widespread adoption of face recognition technologies. This study explores the application of the MobileNetV2-based neural network for human face recognition in authentication systems. The advantages of MobileNetV2, such as its lightweight architecture and high computational efficiency, make it a suitable choice for real-time authentication on edge devices. The proposed method enhances recognition accuracy while maintaining fast processing speeds, ensuring a balance between security and performance. Experimental results demonstrate the effectiveness of the approach under varying lighting conditions and different angles of facial orientation. The study also discusses potential challenges, including spoofing attacks and dataset limitations, and proposes solutions to improve robustness. The findings contribute to the advancement of secure and efficient biometric authentication systems.