Biometrie liveness authentication detection
Sreejit Sundaran, Joycy K. Antony, Krishnan C V Vipin · 2017
Increasing use of biometrie authentication systems in the recent years, spoof fingerprint and iris detection has become increasingly important. Biometrics is to discriminate subjects in a reliable manner for a target application based on one or more signals derived from physical or behavioral traits, such as fingerprint, face, iris, voice, palm, or handwritten sign. Biometric technology has several advantages over common security methods based on some information. In this paper, using of the convolutional neural networks (CNNs) for iris and fingerprint liveness detection is done. The system is evaluated on the data sets used in the liveness detection competition. The comparing of different models: CNNs pre-trained on natural images and fine-tuned with the fingerprint images, CNN with random weights, and a classical local binary pattern(LBP) approach. It is shown that pre-trained CNNs can yield the state-of-the-art results with no need for architecture or hyper-parameter selection. This model has an overall accuracy of 94%.