Fingerprint Presentation Attack Detector Using Global-Local Model

Liu, Haozhe; Zhang, Wentian; Liu, Feng; Wu, Haoqian; Shen, Linlin

doi:10.1109/tcyb.2021.3081764

Cited by 18 publications

(4 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(176). The RTK-PAD method achieved promising results in countering presentation attacks, with a true detection rate (TDR) of 91.19%, an average classification error (ACE) of 2.28%, and a false detection rate (FDR) of 1% (177). Additionally, the OCPAD model and the OCT fingerprint PAD demonstrated efficient spoof detection capabilities using optical coherence technology, achieving a true positive rate (TPR) of 99.43% at a false positive rate (FPR) of 10% and an accuracy of 81.89% with a low error rate of 0.67%, respectively (178; 179).…”

Section: Fingerprint Biometric Authentication Vulnerabilitiesmentioning

confidence: 99%

A Comprehensive Review of Unimodal and Multimodal Fingerprint Biometric Authentication Systems: Fusion, Attacks, and Template Protection

Sumalatha,

Prakasha,

Prabhu

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In the contemporary technological landscape, biometrics, encompassing the analysis of biological data, have gained significance. Biometrics is a methodology that utilize unique behavioral, physical, or morphological traits-such as speech, facial features, iris, fingerprint, retina, and signature for individual identification. Biometric technology has been successfully used in forensic science, security, and authorization systems. This review highlights understanding the classification, types of biometric traits and their comparisons, fingerprint recognition stands out as a reliable and widely adopted method due to its simplicity and cost-effectiveness, accuracy and robustness compared to others. Unimodal fingerprint biometric systems safeguard authentication information through the analysis of characteristic sequences and, face challenges such as vulnerability to spoof attacks, inter-class similarity, intra-class variation, non-universality, and noisy data. These challenges are addressed by multimodal fingerprint biometric systems, in which various biometric sources compensate for each other's limitations. The review focuses on the overview of unimodal and multimodal fingerprint biometric systems and the importance of fusion, advancements in data acquisition, preprocessing, feature extraction, matching algorithms, performance metrics, indexing, template protection, and addressing attacks in enhancing system security and reliability. The review paper sheds light on the intricate relationship between these elements, offering valuable insights into the current state and potential evolution of the field. The review paper highlights current challenges and suggests future research directions, emphasizing the necessity for continual advancements in fusion techniques, template protection methods, and novel defense mechanisms to effectively mitigate emerging threats in unimodal and multimodal fingerprint biometric systems.

show abstract

Section: Fingerprint Biometric Authentication Vulnerabilitiesmentioning

confidence: 99%

A Comprehensive Review of Unimodal and Multimodal Fingerprint Biometric Authentication Systems: Fusion, Attacks, and Template Protection

Sumalatha,

Prakasha,

Prabhu

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The authors in [ 30 ] employed Grad-CAM as a visualization tool to identify and highlight noise across various channels of a network when processing a fingerprint image. The use of CAM is also presented in the study by [ 31 ], where it was used for patch extraction during the inference stage. In our work, we have utilized Grad-CAM not only for visual explanation of the network but also to mitigate the biases present in the dataset through that explanation.…”

Section: Related Workmentioning

confidence: 99%

Enhancing Ensemble Learning Using Explainable CNN for Spoof Fingerprints

Reza,

Jung

2023

Sensors

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have demonstrated remarkable success with great accuracy in classification problems. However, the lack of interpretability of the predictions made by neural networks has raised concerns about the reliability and robustness of CNN-based systems that use a limited amount of training data. In such cases, the utilization of ensemble learning using multiple CNNs has demonstrated the capability to improve the robustness of a network, but the robustness can often have a trade-off with accuracy. In this paper, we propose a novel training method that utilizes a Class Activation Map (CAM) to identify the fingerprint regions that influenced previously trained networks to attain their predictions. The identified regions are concealed during the training of networks with the same architectures, thus enabling the new networks to achieve the same objective from different regions. The resultant networks are then ensembled to ensure that the majority of the fingerprint features are taken into account during classification, resulting in significant enhancement of classification accuracy and robustness across multiple sensors in a consistent and reliable manner. The proposed method is evaluated on LivDet datasets and is able to achieve state-of-the-art accuracy.

show abstract

“…These perturbations are usually small enough to steer by human vision system but can mislead well-performing DNNs models to erroneous predictions with high confidence. This raises great interest and big concern in the CV community, along with the rapid development of adversarial machine learning that focuses on improving the robustnees and interpretability of DNNs [2], [3], [4]. On the contrary, the security threat of DNNs is insufficiently estimated in the field of NLP, especially for some security critical applications, such as spam filtering [5], secret sharing [6], sentiment analysis [7], and webpage phishing [8].…”

Section: Introductionmentioning

confidence: 99%