Hand Gesture-Based Security and Authentication System

In an era marked by growing concerns about data breaches and the limitations of standard authentication systems, this paper introduces an innovative solution: the “Hand Gesture-Based Security and Authentication” system. Recognizing the limitations of passwords and biometric identifiers such as facial recognition, especially in situations when recovery or alteration is impossible, our solution takes advantage of the diversity and durability of hand gestures. Using advanced computer vision and deep learning techniques, we present a system capable of securely authenticating users by deciphering complex hand movements. This unique solution not only reduces the hazards associated with password-based systems, but also tackles the drawbacks of facial recognition, particularly in areas requiring face and hand coverage, such as military, industrial, and hospital settings. This study illustrates the system’s effectiveness in preventing illegal access while maintaining user privacy and adaptability through thorough deployment and evaluation. Our research not only provides a comprehensive answer to today’s authentication difficulties, but it also paves the way for safe access in a variety of operational contexts. The accuracy of $\mathbf{9 5. 9 6 \%}$ is achieved with the ResNet50 model.