ColorVein: Colorful Cancelable Vein Biometrics

TL;DR

ColorVein introduces a colorization-based cancelable vein biometric scheme that converts grayscale vein images into interactive, color-rich representations via hint-point defined color spaces. It combines vein segmentation, a colorization network, and a secure center loss to produce protected templates that preserve discrimination while enabling revocability and unlinkability. Across finger, palm, dorsal hand, and wrist veins, ColorVein delivers competitive recognition performance and robust privacy defenses, including high irreversibility and strong resistance to stolen-token scenarios, with efficient template generation. The approach offers a practical, secure pathway for vein biometrics by enriching feature information through color while providing explicit mechanisms for revocation and cross-database unlinkability.

Abstract

Vein recognition technologies have become one of the primary solutions for high-security identification systems. However, the issue of biometric information leakage can still pose a serious threat to user privacy and anonymity. Currently, there is no cancelable biometric template generation scheme specifically designed for vein biometrics. Therefore, this paper proposes an innovative cancelable vein biometric generation scheme: ColorVein. Unlike previous cancelable template generation schemes, ColorVein does not destroy the original biometric features and introduces additional color information to grayscale vein images. This method significantly enhances the information density of vein images by transforming static grayscale information into dynamically controllable color representations through interactive colorization. ColorVein allows users/administrators to define a controllable pseudo-random color space for grayscale vein images by editing the position, number, and color of hint points, thereby generating protected cancelable templates. Additionally, we propose a new secure center loss to optimize the training process of the protected feature extraction model, effectively increasing the feature distance between enrolled users and any potential impostors. Finally, we evaluate ColorVein's performance on all types of vein biometrics, including recognition performance, unlinkability, irreversibility, and revocability, and conduct security and privacy analyses. ColorVein achieves competitive performance compared with state-of-the-art methods.

Figures (13)

• Figure 1: Vein biometrics: finger vein, palm vein, dorsal hand vein, wrist vein.
• Figure 2: Enrollment/verification procedure of the cancelable biometric system based on the proposed ColorVein scheme.
• Figure 3: Typical vein pattern extraction methods. Row 1 to 4 are finger, palm, dorsal hand and wrist veins. Column 1 to 7 are the original, MC, PC, RLT, GF, IUTW and fusion image, respectively.
• Figure 4: Feature extraction network. $\{I,B\}$ represent an input cancelable vein feature information pairs. $I_X$ and $B_X$ represent enrolled cancelable identities and biometric. $I_Y$ and $B_Y$ represent illegal identities and biometric of impostor. $I_C$ represents cross-application cancelable identities. $B_N$ represents the vein biometrics of the adversary in the stolen scenario.
• Figure 5: Vein image after colorization by the colorization network. Row 1 to 4 are finger, palm, dorsal hand and wrist veins. Columns 1 to 7 indicate the ColorVein hint point numbers of 1, 5, 10, 25, 50, 75 and 100, respectively.