How to assess user interaction effects in Biometric performance

Abstract

Since biometric recognition systems are part of peoples' daily life, manufacturers are more concerned about the data subject interaction and the effects that low usability may have in final users. Why people may not use swipe fingerprint sensors in smartphones? Now that biometric recognition is also used in mobile devices, this influence is higher as the variability of biometrics uses is also higher. The interaction is not fixed anymore, nor scenarios or positions. People unlock their smartphones with the face at night or make payments with the fingertip in a bar. Enrollments are not under control and no proper guides are given to data subjects. All these facts point out the necessity of evaluate the effects of the interaction on the biometric performance as a major factor of influence. This work introduces an evaluation methodology proposed as standard within the ISO/IEC/JTC1/SC37 - Biometrics and its development, justifying the design criteria based on previous experimentation and further works.