|
Activity Number:
|
516
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 2, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statisticians in Defense and National Security
|
| Abstract - #307766 |
|
Title:
|
Statistical Models for Fingerprint Individuality
|
|
Author(s):
|
Sarat Dass*+ and Yongfang Zhu and Anil K. Jain
|
|
Companies:
|
Michigan State University and Michigan State University and Michigan State University
|
|
Address:
|
Department of Statistics and Probability, East Lansing, MI, 48824,
|
|
Keywords:
|
Mixture models ; Clustering ; Biometric Authentication ; Fingerprint Individuality
|
|
Abstract:
|
Forensic evidence based on fingerprints was first challenged in the 1999 case of USA vs. Byron Mitchell, and subsequently, in 20 other cases involving fingerprint evidence. The main concern is the lack of scientific validation of fingerprint evidence. The probability that two different individuals will share a common set of fingerprint features are currently unknown or unsatisfactory. This talk develops a family of finite mixture models to represent the distribution of minutiae in fingerprint images, including minutiae clustering tendencies and dependencies in different regions of the fingerprint image domain. We show that the proposed models better describe the observed variability in the minutiae compared to the uniform model reported in the literature. A mathematical model that computes the probability of a random correspondence is developed based on the mixture models.
|
