Toward Statistically Rigorous Biometric Authentication

• Activity Number: 121
• Type: Contributed
• Date/Time: Monday, August 9, 2004 : 12:00 PM to 1:50 PM
• Sponsor: Biometrics Section
• Author(s): Sinjini Mitra*+
• Companies: Carnegie Mellon University
• Keywords: authentication ; biometrics ; statistical model ; spectrum ; statistical properties

Abstract

The modern world has seen a rapid evolution of the technology of biometric authentication, prompted by an increasing urgency to ensure a system's security. The need for efficient authentication systems has sky-rocketed since 9/11, and the proposed inclusion of digitized photos in passports shows the importance of biometrics in homeland security today. Based on a person's essentially unique biological traits, these methods are potentially more reliable and less prone to fraud than traditional methods like PINs and ID cards. Our goal is to build statistically rigorous authentication tools for handling diverse databases where accuracy is imperative. We first examine statistical properties of an existing system and show how they can make it more attractive to users. We then demonstrate that simple statistical models based on the image spectrum can be used to build authentication schemes by using model coefficients as features. An initial simple model is presented as a natural framework and although it yields low error rates, it ultimately needs to be refined for practical authentication of individuals within a larger population. Several of the associated challenges are discussed.