CBP Biometric Fusion and Multimodal Systems 1 — Questions and Answers
Question 1: Which level of biometric fusion combines raw data or feature vectors from multiple sensors before matching occurs?
- Decision-level fusion
- Score-level fusion
- Feature-level fusion (Correct answer)
- Rank-level fusion
Correct answer: Feature-level fusion
Feature-level fusion operates on the feature vectors extracted from biometric samples prior to matching, merging them into a single combined feature set.
Question 2: What is the primary advantage of using a multimodal biometric system over a unimodal system?
- Lower hardware cost
- Reduced enrollment time
- Improved accuracy and resistance to spoofing (Correct answer)
- Simpler system architecture
Correct answer: Improved accuracy and resistance to spoofing
Multimodal systems combine multiple biometric sources, improving overall accuracy, reducing false accept/reject rates, and making spoofing significantly more difficult.
Question 3: In decision-level fusion, the AND rule requires that:
- At least one matcher accepts the identity claim
- All matchers accept the identity claim (Correct answer)
- The highest matcher score exceeds a threshold
- A weighted average of decisions is computed
Correct answer: All matchers accept the identity claim
The AND rule in decision-level fusion grants access only when every individual matcher independently accepts the claim, maximizing security at the cost of higher false rejection.
Question 4: Which fusion strategy is typically most difficult to implement due to incompatible feature spaces across different biometric modalities?
- Score-level fusion
- Rank-level fusion
- Decision-level fusion
- Feature-level fusion (Correct answer)
Correct answer: Feature-level fusion
Feature-level fusion is the most challenging because feature vectors from different modalities (e.g., fingerprint minutiae and iris codes) often have different dimensionalities and statistical properties.
Question 5: Score normalization is most commonly applied before which type of fusion?
- Feature-level fusion
- Decision-level fusion
- Rank-level fusion
- Score-level fusion (Correct answer)
Correct answer: Score-level fusion
Score-level fusion requires normalization to map match scores from different matchers onto a common scale before combining them, since raw scores from different systems are not directly comparable.
Question 6: A biometric system uses fingerprint and face recognition independently and combines their binary accept/reject outputs. This is an example of:
- Score-level fusion
- Feature-level fusion
- Decision-level fusion (Correct answer)
- Sensor-level fusion
Correct answer: Decision-level fusion
Combining binary accept/reject decisions from independent matchers is the definition of decision-level fusion, the simplest but least flexible fusion approach.
Question 7: What is a 'chimeric' database in the context of multimodal biometric research?
- A database containing spoofed biometric samples
- A database created by virtually pairing samples from different unimodal databases (Correct answer)
- A database of biometric data from multiple countries
- A database encrypted with multiple keys
Correct answer: A database created by virtually pairing samples from different unimodal databases
A chimeric database is an artificially constructed multimodal dataset where multiple unimodal databases are combined by randomly pairing individuals, enabling fusion research without requiring real simultaneous multi-trait capture.
CBP Biometric Fusion and Multimodal Systems 1
Which level of biometric fusion combines raw data or feature vectors from multiple sensors before matching occurs?