Enhancing Security with Face Recognition and Hand Gesture

Traditional authentication techniques like pin and password are widely used but due to its limitations and vulnerabilities, we introduced the biometric authentication techniques like face recognition, fingerprints, iris. But a single layer of authentication system can be prone to data breach. To overcome this, we proposed a dual authentication system using two types of biometrics, which are contact less and non confidential. The combination of face recognition and hand gesture can be used as a dual authentication system. The FaceNet is used for the face recognition and Mediapipe for the hand gesture recognition. Even if the face data is compromised, the hand gesture acts as an additional layer of security. By leveraging the distinctiveness of user defined gestures for various angles, this approach significantly enhances the security and makes spoofing difficult.The proposed model has achieved an accuracy of 90 percentage in recognizing the face and hand gesture.The adaptability of this solution makes it future-ready, aligning with the evolving demands of the modern workplace while ensuring robust protection against unauthorized access. Whenever the system cannot verify either face or hand gesture, the authorization will not be provided. The face embeddings generated by FaceNet of the registered faces and coordinates of the landmark of the hand gesture ensured the privacy of the user information. The proposed system works in a multi threaded environment making it suitable for real time implementation where speed and authentication is necessary. It ensures a reliable verification process.