English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

With the increasing integration of smartphones into our daily lives,fingerphotos are becoming a potential contactless authentication method. Whileit offers convenience, it is also more vulnerable to spoofing using variouspresentation attack instruments (PAI). The contactless fingerprint is anemerging biometric authentication but has not yet been heavily investigated foranti-spoofing. While existing anti-spoofing approaches demonstrated fairresults, they have encountered challenges in terms of universality andscalability to detect any unseen/unknown spoofed samples. To address thisissue, we propose a universal presentation attack detection method forcontactless fingerprints, despite having limited knowledge of presentationattack samples. We generated synthetic contactless fingerprints using StyleGANfrom live finger photos and integrating them to train a semi-supervisedResNet-18 model. A novel joint loss function, combining the Arcface and Centerloss, is introduced with a regularization to balance between the two lossfunctions and minimize the variations within the live samples while enhancingthe inter-class variations between the deepfake and live samples. We alsoconducted a comprehensive comparison of different regularizations' impact onthe joint loss function for presentation attack detection (PAD) and exploredthe performance of a modified ResNet-18 architecture with different activationfunctions (i.e., leaky ReLU and RelU) in conjunction with Arcface and centerloss. Finally, we evaluate the performance of the model using unseen types ofspoof attacks and live data. Our proposed method achieves a Bona FideClassification Error Rate (BPCER) of 0.12\%, an Attack PresentationClassification Error Rate (APCER) of 0.63\%, and an Average ClassificationError Rate (ACER) of 0.37\%.

A Universal Anti-Spoofing Approach for Contactless Fingerprint Biometric Systems