What is it about?

We reveal that current practice of evaluating multimodal biometrics systems can lead to misleading results.

Featured Image

Why is it important?

Multimodal biometrics (fingerprint, face etc) can increase system authentication performance accurately and will be hard for spoof attack. However, exiting evaluation practice has largely used simulation database where each biometric modal belongs to different person. Our work reveals that multimodal biometrics from a person has some correlations and ignoring this could lead to misleading evaluation results.

Perspectives

Traditionally people believe each individual biometric modal is independent and hence simulation multimodal biometrics database has been deployed for evaluation. Our work has shown that this is a wrong practice and such evaluation could lead to misleading evaluation results.

Professor Jiankun Hu
University of New South Wales

Read the Original

This page is a summary of: Mutual dependency of features in multimodal biometric systems, Electronics Letters, February 2015, the Institution of Engineering and Technology (the IET),
DOI: 10.1049/el.2014.4182.
You can read the full text:

Read

Resources

Contributors

The following have contributed to this page