Multi-Scale and Multi-Direction GAN for CNN-Based Single Palm-Vein Identification

Abstract

Deep neural networks have recently achieved promising performance in the vein recognition task and have shown an increasing application trend, however, they are prone to adversarial perturbation attacks by adding imperceptible perturbations to the input, resulting in making incorrect recognition. To address this issue, we propose a novel defense model named MsMemoryGAN, which aims to filter the perturbations from adversarial samples before recognition. First, we design a multiscale autoencoder to achieve high-quality reconstruction and two memory modules to learn the detailed patterns of normal samples at different scales. Second, we investigate a learnable metric in the memory module to retrieve the most relevant memory items to reconstruct the input image. Finally, the perceptional loss is combined with the pixel loss to further enhance the quality of the reconstructed image. During the training phase, the MsMemoryGAN learns to reconstruct the input by merely using fewer prototypical elements of the normal patterns recorded in the memory. At the testing stage, given an adversarial sample, the MsMemoryGAN retrieves its most relevant normal patterns in memory for the reconstruction. Perturbations in the adversarial sample are usually not reconstructed well, resulting in purifying the input from adversarial perturbations. We have conducted extensive experiments on two public vein datasets under different adversarial attack methods to evaluate the performance of the proposed approach. The experimental results show that our approach removes a wide variety of adversarial perturbations, allowing vein classifiers to achieve the highest recognition accuracy.

Extraction of intersecting palm‐vein and palmprint features for cancellable identity verification

“…For example, El-Ghandour et al [20] proposed a palm-vein recognition method based on an enhanced Weber local descriptor (WLD) and stacked autoencoder (SAE) for feature extraction. Qin et al [21] proposed a palm-vein recognition system by image augmentation utilising a Generative Adversarial Network (GAN). They tried to avoid using excessive data for network training by utilising a single image for each identity during training.…”

“…Qin et al. [21] proposed a palm‐vein recognition system by image augmentation utilising a Generative Adversarial Network (GAN). They tried to avoid using excessive data for network training by utilising a single image for each identity during training.…”

Adaptive Deep Feature Fusion for Continuous Authentication With Data Augmentation

“…the conditional Wasserstein GAN, selective VAE, and selective WGAN to augment EEG data for emotion recognition enhancement. The authors in [18] exploited a multiscale and multidirection GAN (MSMDGAN) for data augmentation in a CNN-based single palm-vein identification. In [24], the authors used a conditional Wasserstein GAN (CWGAN) consisting of a CNN-based generator and a CNN-based discriminator for data augmentation in a smartphone-based continuous authentication system.…”

“…These behavioral biometrics-based systems help to secure mobile banking, shopping, remote meeting, and so on, because these applications require to continuously validate the user's identity during an entire session. In this respect, some works utilizing deep learning methods to extract deep features have achieved a relatively high accuracy [16], [17], [18]. These works, however, face the challenge of insufficient training data.…”

Deep Learning-Based Forearm Subcutaneous Veins Segmentation

“…For retinal blood vessel segmentation, Generative Adversarial Networks with U-Net (U-GAN) [20] and Topology Ranking GAN (TR-GAN) [21] have achieved accuracies of 96.15 and 96.29 respectively on DRIVE dataset. For palm vein identification Multi-Scale and Multi-Direction GAN and Convolutional Neural Network (MSMDGAN + CNN) [22] has been proposed, which has produced state of the art results in terms of accuracy. Similarly for finger vein verification Finger vein representation using generative adversarial networks (FV-GAN) [23] based on cycle GAN is proposed which robustly extracts vein patterns and significantly improved equal error rate (EER) and verification accuracy.…”