Presentation Attacks in Palmprint Recognition Systems

Background: A presentation attack places the printed image or displayed video at front of sensor to deceive biometric recognition system. Usually, attackers steal a genuine user’s and use it for attack. In recent years, reconstruction adversarial can generate high-quality fake images, have high success rates. However, their rates degrade remarkably after shooting. Methods: order comprehensively analyze threat palmprint system, this paper makes six datasets. The datasets were tested on texture coding-based methods deep learning-based methods. Results conclusion: experimental results show that caused by leakage original has rate great threat; while decrease significantly.