Deep Learning Approaches for Fingerprint Spoofing Detection Using Visual Data

EasyChair Preprint 13638

Abstract

Fingerprint recognition systems have become a cornerstone of biometric authentication due to their ease of use and reliability. However, they are increasingly vulnerable to spoofing attacks, where artificial replicas of fingerprints can be used to gain unauthorized access. This paper explores the efficacy of deep learning approaches for detecting fingerprint spoofing using visual data.
We begin by providing a comprehensive overview of fingerprint spoofing techniques, including the creation of spoof artifacts using materials such as silicone, gelatin, and latex. These methods present significant challenges for traditional anti-spoofing mechanisms. To address these challenges, we propose leveraging the power of deep learning algorithms, particularly convolutional neural networks (CNNs), which have demonstrated remarkable success in various image classification and pattern recognition tasks.

Keyphrases: Fingerprint, learning approaches, visual data

@booklet{EasyChair:13638,
  author    = {Thomas Micheal},
  title     = {Deep Learning Approaches for Fingerprint Spoofing Detection Using Visual Data},
  howpublished = {EasyChair Preprint 13638},
  year      = {EasyChair, 2024}}