A Universal Anti-Spoofing Approach for Contactless Fingerprint Biometric Systems

View on arXiv ← Back to list

Authors: Banafsheh Adami, Sara Tehranipoor, Nasser Nasrabadi, Nima Karimian

Published: 2023-10-23 15:46:47+00:00

AI Summary

This paper proposes a universal presentation attack detection method for contactless fingerprint biometric systems using a semi-supervised approach. It leverages StyleGAN to generate synthetic spoofed fingerprints, training a modified ResNet-18 model with a novel joint loss function combining Arcface and Center loss for improved accuracy.

Abstract

With the increasing integration of smartphones into our daily lives, fingerphotos are becoming a potential contactless authentication method. While it offers convenience, it is also more vulnerable to spoofing using various presentation attack instruments (PAI). The contactless fingerprint is an emerging biometric authentication but has not yet been heavily investigated for anti-spoofing. While existing anti-spoofing approaches demonstrated fair results, they have encountered challenges in terms of universality and scalability to detect any unseen/unknown spoofed samples. To address this issue, we propose a universal presentation attack detection method for contactless fingerprints, despite having limited knowledge of presentation attack samples. We generated synthetic contactless fingerprints using StyleGAN from live finger photos and integrating them to train a semi-supervised ResNet-18 model. A novel joint loss function, combining the Arcface and Center loss, is introduced with a regularization to balance between the two loss functions and minimize the variations within the live samples while enhancing the inter-class variations between the deepfake and live samples. We also conducted a comprehensive comparison of different regularizations' impact on the joint loss function for presentation attack detection (PAD) and explored the performance of a modified ResNet-18 architecture with different activation functions (i.e., leaky ReLU and RelU) in conjunction with Arcface and center loss. Finally, we evaluate the performance of the model using unseen types of spoof attacks and live data. Our proposed method achieves a Bona Fide Classification Error Rate (BPCER) of 0.12%, an Attack Presentation Classification Error Rate (APCER) of 0.63%, and an Average Classification Error Rate (ACER) of 0.37%.

Key findings

The proposed method achieved a Bona Fide Classification Error Rate (BPCER) of 0.12%, an Attack Presentation Classification Error Rate (APCER) of 0.63%, and an Average Classification Error Rate (ACER) of 0.37%. The model showed significant improvements over existing methods, particularly in detecting unseen spoof attacks.

Approach

The authors generate synthetic contactless fingerprints from live data using StyleGAN. These synthetic samples, along with real live fingerprints, are used to train a semi-supervised ResNet-18 model using a novel joint loss function that combines Arcface and Center loss with regularization to balance inter- and intra-class variations.

Datasets

CLARKSON and COLFISPOOF datasets

Model(s)

Modified ResNet-18

Author countries

USA

← Previous