Name : Stephanie Schuckers

Institution : Clarkson University

Stephanie Schuckers is an associate professor in the Department of Electrical and Computer Engineering at Clarkson University. Schuckers received the B.S. in electrical engineering from the University of Iowa in 1992. As a Whitaker Foundation Graduate Fellow, she received the M.S. and Ph.D. degree in electrical engineering from the University of Michigan in 1994 and 1997, respectively. Her research focuses on processing and interpreting signals which arise from the human body.  Signals include the electrocardiogram, biometric signals like fingerprints, respiration, and electroencephalograms.  Methods involve classic signal processing, statistical techniques, pattern recognition, algorithm development and evaluation, data mining, and image processing. Much of her work involves analysis of real data collected from human, cadaver, and animal studies. For example,in work with the Department of Homeland Security, she is studying methods to increase the security of biometric systems, like fingerprint, iris, and face. Her work is funded from various sources, including National Science Foundation, American Heart Association, National Institute of Health, Department of Homeland Security, the Center for Identification Technology, and private industry, among others.

Publications : Aditya Abhyankar, Stephanie Schuckers Integrating a wavelet based perspiration liveness check with fingerprint recognition, Pattern Recognition, vol. 42, pp. 452-464, March 2009.

Derakhshani R, Schuckers SAC, Determination of Vitality From A Non-Invasive Biomedical Measurement for Use in Fingerprint Scanners, Pattern Recognition, No.2 pp. 383-396, 2003.

Title of Project : Reduction of Vulnerability of Fingerprint Biometric Authentication Systems to Spoofing

There is a need to tie an individual to a specific transaction including such applications as border security, authentication, identity fraud, etc. Biometrics, which relies of the physiologic or behavior characteristics of an individual, has the property to tie an individual to a specific time and place for identity management. However, the promise of biometrics has been weakened as it has been shown that it is not difficult to make molds of latent fingerprints left by legitimate users or stolen from a database in order to create fake finger replicas, or ‘spoofs’, made from Play-Doh, gelatin, silicone and other materials to fool a variety of fingerprint scanners. Even though biometric devices use physiologic information for authentication purposes, these measurements rarely indicate ‘liveness’ of the information presented. The goal of liveness testing is to determine if the biometric being captured is an actual measurement from the authorized, live person who is present at the time of capture. To quote Dorothy Denning, “it is ‘liveness’, not secrecy, that counts” as a key factor in biometric-based identity management systems, particularly for unsupervised applications. Previously, our laboratory has demonstrated that, unlike spoof and cadaver fingers, live fingers demonstrate a distinctive spatial moisture pattern when in physical contact with the capturing surface of the fingerprint scanner. Image/signal processing and pattern recognition algorithms have been developed to quantify this phenomenon using wavelet and statistical approaches. Recent results have assessed this method on a larger dataset than previously considered which contains 4000 live fingerprints (81 subjects with 2 fingers for an average of 4 sessions) and 4000 spoof fingerprints (made from Play-Doh, gelatin and silicone molds). Average results for cross validation indicate an equal error rate (equal error % of spoof detected as live and live detected as spoof) of 3.5%. Additional research is ongoing to expand the types of spoof attacks, the impact of environmental conditions on the liveness algorithm, and addition of features to further improve performance. Liveness detection is a critical component to achieve the full power of biometric systems in order to achieve identity management in complex applications such as border crossing and authentication systems. Through liveness detection, we can minimize security risks associated with biometrics while simultaneously utilizing the benefits of biometrics which ties an individual to a specific transaction.