Fingerprint liveness detection based on multi-modal fine-grained feature fusion

Abstract

Recently, fingerprint recognition systems are widely deployed in our daily life. However, spoofing via using special materials such as silica, gelatin, Play-Doh, clay, etc., is one of the most common methods of attacking fingerprint recognition systems. To handle the above defects, a fingerprint liveness detection (FLD) technique is proposed. In this paper, we propose a novel structure to discriminate genuine or fake fingerprints. First, to describe the subtle differences between them and make full use of each algorithm, this paper extracts three types of different fine-grained texture features, such as SIFT, LBP, HOG. Next, we developed a feature fusion rule, including five fusion operations, to better integrate the above features. Finally, those fused features are fed into an SVM classifier for the subsequent classification. Experimental results on the benchmark LivDet 2013 fingerprints indicate that the classification performance of our method outperforms other FLD methods proposed in recent literature.