Image Biometric Verification in Spatial Frequency Domain

Prof. Vijayakumar Bhagavatula

Dept. of Electrical and Computer Engineering & CyLab, Carnegie Mellon University, Pittsburgh, USA

Date and time: 11.30am - 12.30pm, Friday 6 October, 2006

Venue: 12.05.02 (Building 12, Level 5, Room 2)

Chair: Dr. Lawrence Cavedon

Abstract:

Verifying a user’s identity is critical in many applications such as access control to physical and virtual spaces. Most current authentication systems are password-based or based on the use of swipe cards or RF ID tags. These authentication approaches based on “what you know” and “what you have” are susceptible to problems such as forgetting the password and cards being lost or stolen. One way to overcome these problems is to employ biometrics (e.g., fingerprints, face images, signatures, etc.) for authentication since they rely on “what you are”. Biometrics are physiological or behavioral characteristics that are unique to an individual. Some biometric modalities produce images and biometric verification involves matching these images. Most image biometric matching methods work in image domain, but there are advantages to working in the spatial frequency domain. This talk will provide an overview of spatial frequency domain image processing methods (simply called advanced correlation filters) to authenticate a person’s identity based on their biometrics. In particular, the application of correlation filters to verify the identity based on face images and iris images will be discussed. Correlation filters offer several advantages such as shift-invariance, closed-form designs and graceful degradation. Although the focus of this talk is on verification, we will also show results of applying these methods to the task of face identification. We will also discuss the performance of correlation filters in independent evaluations such as the face recognition grand challenge (FRGC) and the iris challenge evaluation (ICE).

About the speaker:

Vijayakumar Bhagavatula received his Ph.D. in Electrical Engineering from Carnegie Mellon University (CMU), Pittsburgh and since 1982, he has been a faculty member in the Electrical and Computer Engineering (ECE) Department at CMU where he is now a Professor. He served as the Associate Head of the ECE Department from 1994 to 1996 and as its Acting Department Head during 2004-2005. In 2003, he received the Eta Kappa Nu Most Outstanding Teacher award in ECE Department at Carnegie Mellon University. Professor Kumar's research interests include Pattern Recognition (for automatic target recognition and biometrics applications) and Coding and Signal Processing for Data Storage Systems and for Digital Communications. He has authored or co-authored more than 400 technical papers and one book entitled Correlation Pattern Recognition. He served as a Topical Editor for the Information Processing division of Applied Optics. Professor Kumar is a member of Sigma Xi, a senior member of IEEE, a Fellow of SPIE - The International Society of Optical Engineering, a Fellow of Optical Society of America (OSA) and a Fellow of the Intl. Association of Pattern Recognition (IAPR). He is listed in Marquis’ Who’s Who in the World and in the American Men and Women of Sciences.

---

Seminar Organisation

Seminars are free and open to the general public. No booking is necessary. If you are interested in giving a presentation in this seminar series, or to make suggestions for speakers, please contact Xiaodong Li, the seminar co-ordinator.