Extreme Populations

This research is being led by Dr. Elliott

Our group focuses on the data collection of "extreme populations", that may not be available to all researchers. Two examples are the elderly and those that have illnesses that can affect a biometric either through the illness or treatment. The reason for this research is simple: for biometrics to become pervasive in society, we need to understand the performance of biometrics across all segments of society, not just those that are easy to collect.

So the challenges are several, and are multidisciplinary in nature. For example, as we get older, our biometric traits may change over time. A study conducted in 2003-2004 examined fingerprint image quality over two different populations (18-25 year olds and greater than 65). Further research is being undertaken this semester to see whether there is any change in the performance of a fingerprint matcher across different levels of fingerprint quality.

Other research that is being undertaken relates to voice recognition, and its performance when the subject is sick, or has a sore throat. The lab is also undertaking a survey on perceptions of the elderly towards biometric devices. Results of the survey will appear around April 2006.

Previous Work and Publications

	Sickler*, N. C., & Elliott, S. J. (2005). An evaluation of fingerprint image quality across an elderly population vis-a-vis an 18-25 year old population. Proceedings of the 39th Annual International Carnahan Conference on Security Technology (ICCST) (pp. 61-68). Las Palmas de G. C., Spain.
	Sicklern, N. C., (2004). An evaluation of fingerprint image quality vis-a-vis an 18-25 year old population baseline, Biometrics Consortium, Washington, DC.