Secure Gait Recognition-Based Smart Surveillance Systems Against Universal Adversarial Attacks

Currently, the internet of everything (IoE) enabled smart surveillance systems are widely used in various fields to prevent forms abnormal behaviors. The authors assess vulnerability based on human gait and suggest a defense strategy secure them. Human recognition is promising biometric technology, but one significantly hindered because universal adversarial perturbation (UAP) that may trigger system failure. More specifically, this research study, emphasize sample convolutional neural network (CNN) model design for its susceptibility UAPs. compute as non-targeted UAPs, which failure lead an inaccurate label input given subject. findings show analysis susceptible even if norm generated noise substantially less than average images. Later, next stage, illustrate mechanism gait.