Authors of Keynote Papers: Akira Otsuka

Conventionally, almost all biometric authentication systems assume that impersonation attempts are conducted by human samples or artifacts resemble to human samples. Similarly, the security of biometric authentication systems is conventionally measured by false acceptance rate, the average probability of accepting impersonation attempts assuming biometric samples are uniformly chosen from human bodies. Obviously, however, adversaries are not limited to use samples from human. Deep analysis of biometric authentication algorithms often shows that some irregular artificial patterns give very high acceptance rate against any registered human templates. Wolf attack is defined as a series of attacks utilizing such patterns in impersonation against biometric authentication systems. In this talk, I will demonstrate our recent results in fingerprint recognition systems and theoretical and practical countermeasures against it.