DEPARTMENT OF COMPUTER SCIENCE COLLOQUIUM SERIES

Anil K. Jain
Department of Computer Science and Engineering
Michigan State University, East Lansing

Biometric Recognition: How Do I Know Who You Are?
Friday, April 16, 2010, 12:00pm, room 70-1610


ABSTRACT
A number of government and business units require accurate and
robust personal recognition methods to either confirm or determine
the identity of an individual requesting their services. The purpose
is to ensure that only an authorized user, and not anyone else,
avails the rendered services. Examples of such applications include
(i) access to secure facility, (ii) computer system or network access,
(iii) money withdrawal from ATM, (iv) obtaining official documents
(driver license), and (v) receiving government services such as
welfare disbursements.

Biometric recognition, or simply biometrics, refers to automatic
recognition of an individual based on his anatomical or behavioral
characteristics (fingerprint, face, voice). Biometrics allows us
to confirm or establish an individual's identity based on who she
is, rather than by what she possesses (ID card) or what she remembers
(PIN). As a result, biometrics offers many opportunities to enhance
security and reduce financial fraud. Biometric systems also introduce
an aspect of user convenience; they alleviate the need for a user
to remember multiple passwords associated with different applications.
The most well known example of a biometric system is the Automatic
Fingerprint Identification System (AFIS) used by major law enforcement
agencies worldwide.

Biometric systems now are being increasingly adopted in many large
scale civilian applications such as national ID cards and e-passports
as well as personal devices such as laptops. In spite of success in
identity management, biometric systems still face many challenges:
non-universality of a trait (failure to enroll), noisy and distorted
biometric data, finite error rate, and spoof attacks (system security).
Multimodal biometric systems that fuse evidence presented by multiple
identity traits (e.g., face and iris) can alleviate some of the problems.
We will present a brief overview of biometrics and challenges that need
the attention of pattern recognition, image processing and computer
vision researchers.


BIOGRAPHY
Anil Jain is a University Distinguished Professor in the Department
of Computer Science at Michigan State University. His research interests
include pattern recognition, computer vision, machine learning and
biometric recognition. He received the K.S. Fu Prize and Guggenheim,
Humboldt, Fulbright, IEEE Computer Society Technical Achievement and
IEEE Wallace-McDowell awards. He is a Fellow of ACM, IEEE, SPIE, IAPR,
and AAAS and served as the Editor-in-Chief of the IEEE Transactions
on Pattern Analysis and Machine Intelligence (1991-94). Holder of six
patents in the area of fingerprints, he is the co-author of a number
of books, including Handbook of Biometrics (2007), Handbook of
Multibiometrics (2006), Handbook of Face Recognition (2005), Handbook
of Fingerprint Recognition (2003) and Algorithms for Clustering Data
(1988). He served on The U.S. National Academies committees on Whither
Biometrics and Improvised Explosive Devices and the U.S. Defense
Science Board.