CBP Biometric Performance Evaluation and Testing Flashcards

What does FMR (False Match Rate) measure in a biometric system?

Answer: The rate at which impostors are incorrectly accepted as genuine

FMR (False Match Rate), also known as False Accept Rate (FAR), measures the frequency at which an impostor's biometric sample is incorrectly matched to another person's enrolled template.

What is the FNMR (False Non-Match Rate) in biometric system evaluation?

Answer: The probability that a genuine user's sample fails to match their enrolled template

FNMR (False Non-Match Rate) is the probability that a genuine user's presented biometric sample fails to match their stored template, causing a false rejection.

On an ROC (Receiver Operating Characteristic) curve for a biometric system, what trade-off is being visualized?

Answer: FMR vs. FNMR across different decision thresholds

A biometric ROC curve plots FMR against FNMR (or 1-FNMR) at varying decision thresholds, showing the trade-off between security and user convenience.

The Equal Error Rate (EER) is the point on the ROC curve where:

Answer: FMR equals FNMR

The Equal Error Rate (EER) is the operating point where FMR equals FNMR, and is commonly used as a single-figure summary of biometric system accuracy.

What does the 'Failure to Enroll' (FTE) rate measure?

Answer: The proportion of individuals whose biometric samples cannot be successfully enrolled

FTE (Failure to Enroll) rate measures the proportion of users for whom the system cannot create a usable biometric template during the enrollment process.

In biometric testing, what is a 'technology evaluation' as opposed to an 'operational evaluation'?

Answer: Technology evaluation tests algorithms on archived data; operational evaluation tests complete systems in real deployment conditions

A technology evaluation tests biometric algorithms on archived datasets in controlled conditions, while an operational evaluation tests the complete system in a real or simulated live deployment environment.