Biometrics

Biometrics

1. [10 points] In many applications, traditional authentication mechanisms based on ID Cards
(token-based authentication) and Passwords (knowledge-based authentication) are used to
verify the identity of a user. Based on O’Gorman’s paper, describe the advantages and disadvantages
of traditional authentication schemes compared to a biometric-based authentication
system.
2. [10 points] Let B1
, B2 and B3 denote 3 different fingerprint matchers that are used to generate
genuine and impostor match scores on a fixed set of fingerprint images. The mean (µ) and
variance (s
2
) of the genuine and impostor score distributions resulting from the 3 different
matchers are tabulated below.
Matcher Genuine Impostor
µ s
2 µ s
2
B1 10 25 60 35
B2 60 5 80 10
B3 40 15 70 15
Based on the score statistics, determine which one of the three matchers has performed well
and which one has performed the worst. Provide adequate numerical justification.
3. [10 points] Consider an experiment in which you are provided the face images of 7 subjects.
The number of images collected from each subject is tabulated below:
Subject Number Number of Images
001 10
002 8
003 1
004 5
005 8
006 2
007 3
1
Based on these numbers, what is the total number of genuine and impostor scores that can be
generated using a symmetric face matcher? Explain your answer.
4. [10 points] Consider a scenario wherein a fingerprint-based biometric system is installed in
a grocery store in East Lansing. Assume that shoppers have the option of enrolling into
the system. This would permit them to render payment at the checkout register by merely
placing their index finger on a fingerprint sensor and typing in a 4-digit PIN. After successfully
verifying the shopper’s identity, the system would then connect to their bank account and debit
the amount of the purchase.
Based on the terminology developed in class, explain how you would characterize this biometric
system (also see Section 1.5.1 in the text book). You must justify your answer with a
detailed explanation.
5. [10 points] Consider a biometric matcher that generates similarity scores in the range [0,
1]. Its genuine and impostor score distributions are as follows: p(s|genuine) = 2s and
p(s|impostor) = 2 – 2s. Suppose the following decision rule is employed: s is classified
as a genuine score if s = ?; else it is classified as an impostor score. Here, ? ? [0, 1].
• Plot the two distributions in a single graph.
• If ? = 0.5, what is the FMR (i.e., FAR) and FNMR (i.e., FRR) of the biometric matcher?
• If ? = 0.6, what is the FMR (i.e., FAR) and FNMR (i.e., FRR) of the biometric matcher?
• Plot the ROC curve based on these distributions.
2
FOR YOUR ASSIGNMENTS TO BE DONE AT A CHEAPER PRICE PLACE THIS ORDER OR A SIMILAR ORDER WITH US NOW