CFVNet: An End-to-End Cancelable Finger Vein Network for Recognition

TL;DR

This work proposes an end-to-end cancelable finger vein network (CFVNet), which can be used to design an secure finger vein recognition system and integrates preprocessing and template protection into an integrated deep learning model.

Abstract

Finger vein recognition technology has become one of the primary solutions for high-security identification systems. However, it still has information leakage problems, which seriously jeopardizes users privacy and anonymity and cause great security risks. In addition, there is no work to consider a fully integrated secure finger vein recognition system. So, different from the previous systems, we integrate preprocessing and template protection into an integrated deep learning model. We propose an end-to-end cancelable finger vein network (CFVNet), which can be used to design an secure finger vein recognition system.It includes a plug-and-play BWR-ROIAlign unit, which consists of three sub-modules: Localization, Compression and Transformation. The localization module achieves automated localization of stable and unique finger vein ROI. The compression module losslessly removes spatial and channel redundancies. The transformation module uses the proposed BWR method to introduce unlinkability, irreversibility and revocability to the system. BWR-ROIAlign can directly plug into the model to introduce the above features for DCNN-based finger vein recognition systems. We perform extensive experiments on four public datasets to study the performance and cancelable biometric attributes of the CFVNet-based recognition system. The average accuracy, EERs and Dsys on the four datasets are 99.82%, 0.01% and 0.025, respectively, and achieves competitive performance compared with the state-of-the-arts.

Figures (11)

• Figure 1: Cancelable finger vein network (CFVNet) overall architecture. Localization, Compression and Transformation represent the three sub-functional modules of the BWR-ROIAlign. De-R Conv is the de-redundant convolution. External key $K$ is the pseudo-random token generated during user registration.
• Figure 2: ROI localization sub-module.
• Figure 3: Spatial redundancy removal without quantization loss.① ROI coordinates mapped to corresponding grid cells on the feature map, ② divide the mapped grid cells into subregions of $H\times W$ number, ③ Using linear interpolation in coordinate mapping and meshing.
• Figure 4: Redundant feature map visualization. The green boxes contain little feature information, while the red boxes contain almost no feature information.
• Figure 5: De-R Conv Architecture. Here, $\mathbf{\textbf{+}}$ represents the positions summed.