An introduction to biometric-completeness: the equivalence of matching and quality

Authors:
P. Jonathon Phillips;J. Ross Beveridge
Affiliations:
National Institute of Standards and Technology, Gaithersburg, MD;Department of Computer Science, Colorado State University, Fort Collins, CO
Venue:
BTAS'09 Proceedings of the 3rd IEEE international conference on Biometrics: Theory, applications and systems
Year:
2009

Citing 4
Cited 0

Performance of Biometric Quality Measures

IEEE Transactions on Pattern Analysis and Machine Intelligence
The Traveling Salesman Problem: A Computational Study (Princeton Series in Applied Mathematics)

The Traveling Salesman Problem: A Computational Study (Princeton Series in Applied Mathematics)
Factors that influence algorithm performance in the Face Recognition Grand Challenge

Computer Vision and Image Understanding
How features of the human face affect recognition: a statistical comparison of three face recognition algorithms

CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper introduces the concept of biometric-completeness. A problem is biometric-complete if solving the problem is "equivalent" to solving a biometric recognition problem. The concept of biometric-completeness is modeled on the informal concept of artificial intelligence (AI) completeness. The concept of biometric-completeness is illustrated by showing a formal equivalence between biometric recognition and quality assessment of biometric samples. The model allows for the inclusion of quality of biometric samples in verification decisions. The model includes most methods for incorporating quality into biometric systems. The key result in this paper shows that finding the perfect quality measure for any algorithm is equivalent to finding the perfect verification algorithm. Two results that follow from the main result are: finding the perfect quality measure is equivalent to solving the open-set and closed-set identification problems; and that a universal perfect quality measure cannot exist.