End-to-End Deep Learning Fusion of Fingerprint and Electrocardiogram Signals for Presentation Attack Detection

Abstract

Although fingerprint-based systems are the commonly used biometric systems, they suffer from a critical vulnerability to a presentation attack (PA). Therefore, several approaches based on a fingerprint biometrics have been developed to increase the robustness against a PA. We propose an alternative approach based on the combination of fingerprint and electrocardiogram (ECG) signals. An ECG signal has advantageous characteristics that prevent the replication. Combining a fingerprint with an ECG signal is a potentially interesting solution to reduce the impact of PAs in biometric systems. We also propose a novel end-to-end deep learning-based fusion neural architecture between a fingerprint and an ECG signal to improve PA detection in fingerprint biometrics. Our model uses state-of-the-art EfficientNets for generating a fingerprint feature representation. For the ECG, we investigate three different architectures based on fully-connected layers (FC), a 1D-convolutional neural network (1D-CNN), and a 2D-convolutional neural network (2D-CNN). The 2D-CNN converts the ECG signals into an image and uses inverted Mobilenet-v2 layers for feature generation. We evaluated the method on a multimodal dataset, that is, a customized fusion of the LivDet 2015 fingerprint dataset and ECG data from real subjects. Experimental results reveal that this architecture yields a better average classification accuracy compared to a single fingerprint modality.

A Multitier Deep Learning Model for Arrhythmia Detection

“…To overcome the previous ML problems, researchers used deep learning approaches. Recently, deep learning is widely used in many fields [30][31][32], especially in medical fields [33][34][35]. For arrhythmia detection, several methods are presented [20][21][22][23][24][25][26][27][28][29].…”

Deep Feature Fusion of Fingerprint and Online Signature for Multimodal Biometrics

“…Similarly, Surendra et al [29] developed a biometric based attendance system using a deep learning-based feature fusion technique on the iris features. The features from fingerprints and electrocardiogram (ECG) were fused by Jomaa et al [30] to detect the presentation attacks. They deployed three CNN architectures including fully connected layers, 1-D CNN and 2-D CNN for the feature extraction from ECG and an EfficientNet for the feature extraction from the fingerprint.…”

Transformers and Generative Adversarial Networks for Liveness Detection in Multitarget Fingerprint Sensors

“…Cycle consistency loss compares a source image to the generated target image and calculates the difference between them. We should arrive where we started when we translate from one domain to another and back again, F(G(X)) ≈ X and G(F(X)) ≈ X [15]. The cycle consistency loss is:…”

“…In a recent work, a new fingerprint PAD method based on the fusion of fingerprint with a more secure biometric modality (such as the face, ECG, and fingerprint dynamics) were proposed in the literature. M. Jomaa et al [15] proposed an end-to-end deep learning-based fusion neural architecture between a fingerprint and electrocardiogram (ECG) signals. The proposed method uses EfficientNets for fusing ECG and fingerprint feature representations.…”