Abstract | ||
---|---|---|
Attacking iris systems with fake iris patterns has become the largest security risk of iris recognition systems. Therefore iris liveness detection which discriminate genuine or fake iris images is of significant importance to iris recognition systems. However, the state-of-the-art algorithms mainly rely on hand-crafted texture features which can only identify fake iris images with single pattern. This paper proposes a Multi-patch Convolution Neural Network (MCNN) that is capable of handling different types of fake iris images. MCNN directly learns the mapping function between raw pixels of the input iris patch and the labels. The outputs of each patch are fed into a decision layer which determines the final decision. Our proposed algorithm automatically learns the features to detect hybrid pattern of fake iris images rather than handcraft. The decision layer helps to improve the robustness and accuracy for iris liveness detection. Experimental results demonstrate an extremely higher accuracy of iris liveness detection than other state-of-the-art algorithms. The proposed MCNN remarkably achieve the best results with nearly 100% accuracy on ND-Contact and CAISA-Iris-Fake datasets. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/BTAS.2016.7791186 | 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS) |
Keywords | Field | DocType |
multipatch convolution neural network,iris liveness detection,iris recognition systems,fake iris images,mapping function,decision layer | Computer vision,Iris recognition,Algorithm design,Pattern recognition,Convolutional neural network,Computer science,Feature extraction,Robustness (computer science),Pixel,Artificial intelligence,Artificial neural network,Liveness | Conference |
ISSN | ISBN | Citations |
2474-9680 | 978-1-4673-9734-6 | 3 |
PageRank | References | Authors |
0.38 | 0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lingxiao He | 1 | 13 | 4.24 |
Haiqing Li | 2 | 77 | 7.57 |
Fei Liu | 3 | 110 | 15.24 |
Nianfeng Liu | 4 | 10 | 0.84 |
Zhenan Sun | 5 | 2379 | 139.49 |
Zhaofeng He | 6 | 135 | 8.28 |