Detection of Fake Biometrics - Assessment of Image Quality in Face, Fingerprint

Abstract

Biometric authentication is difficult because it is difficult to distinguish between a genuine trait and a fake, self-made synthetic, or reconstructed sample. This calls for new, more effective protection measures. This paper describes a software-based fake detection method that can be used in multiple biometric systems to detect fraudulent access attempts. This system adds liveness assessment to biometric recognition systems in an easy-to-use, non-invasive way. It is intended to increase security. This method is easy to use and can be used in real-time. It works by combining 25 image quality features (i.e. it uses 25 general image quality features from one image (i.e., the same image used for authentication) to differentiate legitimate samples from counterfeit ones. The proposed method is superior to other state-of-the-art methods, according to experiments that used publicly available data for fingerprint, iris, and 2D faces. The analysis of biometric samples' general quality revealed valuable information that can be used to identify genuine traits from fake ones.