Biometrics: your face, your fingerprint, your voice
June 13, 2011 ·
The science of biometrics involves methods for recognizing and identifying people based on fundamental physical or behavioral traits.The oldest and most familiar use is fingerprinting.
At West Virginia University the biometrics program is part of the Lane Department of Computer Science and Electrical Engineering. Dr. Bojan Cukic is Co-Director of the Center for Identification Technology Research at WVU.
"The science of biometrics is relatively new probable just about a hundred years old. In many environments, especially in the forensics investigations, law enforcement agencies wanted to know who the person who committed the crimes were and how to find them. we don’t have paper cards anymore in the filing systems, we have data bases and we are also able to use computers, cameras, and other types of sensors to deposit or capture human iris, appearance of the human face or just the appearance of the fingerprint and use those modalities to search through very large data bases very accurately and quickly," said Dr. Cukic.
Dr. Cukic says, although the science behind biometrics has been evolving for years, there have been major strides during the past 20-25 years. This is due in part to more powerful computers and increasingly high quality digital cameras.
Dr. Arun Ross is an Associate Professor at WVU, working in the Center for Identification Technology Research.
"The examples include fingerprint recognition, iris recognition, speaker recognition, face recognition, hand geometry recognition, and so forth. So what you’re really doing is you’re moving past traditional authentication schemes, like passwords, ID cards, driver’s licenses, passports where identity is tied to a document, which is what you possess or to knowledge, like a password which is what you remember. So you’re moving away from token based recognition or knowledge based recognition to something that is based on an intrinsic property of that individual," Dr. Ross said.
Dr. Cukic says WVU’s involvement in biometrics is relatively new but he cites a major event that helped raise awareness among the general public.
"In the middle of [1990s] you would remember the O. J. Simpson case, which played out in courts in L.A., and at that time one of the major problems that law enforcement agencies had was having experts who were able to talk in very simple terms to the juries and convince them that the evidence is actually strong," Dr. Cukic said. "At that time, in 1997, West Virginia University and the FBI came up with a memorandum of understanding following which West Virginia University developed two academic programs. One academic program was in forensics sciences and the second academic program was in biometric systems."
Dr. Cukic and his colleagues at the Center for Identification Technology Research are looking toward the future of biometric identification.
"So we are analyzing how to recognize the person at night at the long distance. We are analyzing different scenarios, for example very low quality DNA sampling and DNA recognition which means if there is a very degraded sample of DNA that may have been left at the crime scene, can we at least say with some probability that this is the perpetrator of that crime, rather than being able to say with very very high probability or almost certainty, yes this is the person," Dr. Cukic said.
Dr. Cukic and Dr. Ross expect advances in the area of Identification Technology to continue accelerating at a rapid pace.