We consider quantitatively establishing the discriminative power of iris biometric data. It is difficult, however, to establish that any biometric modality is capable of distinguishing every person because the classification task has an extremely large and unspecified number of classes. Here, we propose a methodology to establish a measure of discrimination that is statistically inferable. To establish the inherent distinctness of the classes, i.e., to validate individuality, we transform the many class problem into a dichotomy by using a distance measure between two samples of the same class and between those of two different classes. Various features, distance measures, and classifiers are evaluated. For feature extraction we compare simple binary and multi-level 2D wavelet features. For distance measures we examine scalar distances, feature vector distances, and histogram distances. Finally, for the classifiers we compare Bayes decision rule, nearest neighbor, artificial neural network, and support vector machines. Of the eleven different combinations tested, the best one uses multi-level 2D wavelet features, the his-togram distance, and a support vector machine classifier.

Keywords: Biometric individuality, Dichotomy, Iris.

Biography:

Sung-Hyuk Cha was born and grew up in Seoul, Korea and received his Ph.D. in Computer Science from State University of New York at Buffalo in 2001 and B.S. and M.S. degrees in Computer Science from Rutgers, the State University of New Jersey in 1994 and 1996, respectively. During his undergraduate years, he became a member of Phi Beta Kappa and Golden Key National Honor Society. He graduated with High Honors and received High Honors in Computer Science. During his Master's years, he developed "Fast Image Template and Dictionary Matching Algorithms" under guidance of Prof. Martin Farach. From 1996 to 1998, he was working in the area of medical information systems such as PACS, teleradiology, and telemedicine at Information Technology R&D Center, Samsung SDS. During his PhD years, he was affiliated with the Center of Excellence for Document Analysis and Recognition (CEDAR). Major contribution made at CEDAR includes dichotomy model to establish the individuality of handwriting, distance measures on histograms and strings, a nearest neighbor search algorithm, apriori algorithm, etc. supervised by Prof. Sargur N. Srihari. He has been a faculty member of Computer Science department at Pace University since 2001. His main interests include computer vision, data mining, pattern matching & recognition. He is a member of AAAI, IEEE and its Computer Society and IS&T.