Liveness Verification for Face Recognition System

Computer Engineering

The popularity of face recognition is increasing due to its ability to recognize human faces, making it a popular topic for computer vision research. Despite its usefulness in image analysis and understanding, face recognition systems face challenges with fraudulent identities, commonly known as spoofing attacks. To address this challenge, face liveness detection is an important biometric authentication method that distinguishes between authentic and fake faces. Liveness detection can be implemented using software or hardware means. However, the use of additional hardware can be costly. The project aims to develop a liveness verification system for face recognition that can accurately distinguish between live and spoofed faces. Spoofing attacks on face recognition systems are becoming increasingly common, and therefore, it is necessary to develop robust liveness verification techniques. The proposed system employs a combination of physiological and behavioral biometric modalities to detect fake faces, including eye blinking, head movement, and facial expressions. The system also utilizes machine learning algorithms to analyze and classify live and spoofed faces based on various features extracted from the biometric data. The effectiveness of the proposed system is evaluated using a large dataset of live and spoofed faces, and the results show a high accuracy in detecting fake faces. The developed liveness verification system has the potential to improve the security of face recognition systems and prevent unauthorized access to sensitive information. Therefore, this paper proposes a CNN-based model for liveness verification and a web-based application to demonstrate its real-time implementation.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.