A study of liveness detection in fingerprint and iris recognition systems using image quality assessment

Biometric recognition systems are vulnerable to diverse attacks that emerged as a challenge in adopting these systems in real life scenarios. Development of efficient security measures is a necessity to ensure the actual presence of a genuine or real trait in various biometric recognition systems. This study aims at Liveness Detection in Fingerprint and Iris Recognition systems using Image Quality Assessment. The potential of general image quality assessment as a protection method against different biometric attacks is explored here. The key idea of this approach is to present a software based multi-biometric and multi attack protection method that characterize the real traits. The proposed method extracts 25 image quality assessment features from a single input image to build an appropriate classifier, which classifies the test image as real or fake given the extracted set of features.