Multimodal Biometric Template Protection Based on a Cancelable SoftmaxOut Fusion Network

Abstract

Authentication systems that employ biometrics are commonplace, as they offer a convenient means of authenticating an individual’s identity. However, these systems give rise to concerns about security and privacy due to insecure template management. As a remedy, biometric template protection (BTP) has been developed. Cancelable biometrics is a non-invertible form of BTP in which the templates are changeable. This paper proposes a deep-learning-based end-to-end multimodal cancelable biometrics scheme called cancelable SoftmaxOut fusion network (CSMoFN). By end-to-end, we mean a model that receives raw biometric data as input and produces a protected template as output. CSMoFN combines two biometric traits, the face and the periocular region, and is composed of three modules: a feature extraction and fusion module, a permutation SoftmaxOut transformation module, and a multiplication-diagonal compression module. The first module carries out feature extraction and fusion, while the second and third are responsible for the hashing of fused features and compression. In addition, our network is equipped with dual template-changeability mechanisms with user-specific seeded permutation and binary random projection. CSMoFN is trained by minimizing the ArcFace loss and the pairwise angular loss. We evaluate the network, using six face–periocular multimodal datasets, in terms of its verification performance, unlinkability, revocability, and non-invertibility.

Feature extraction and learning approaches for cancellable biometrics: A survey

“…ResNet-50 has 50 layers, including three types of layers, namely convolutional layers, activation layers and batch normalisation layers. Kim et al [97] investigated an end-to-end multimodal cancellable biometric scheme using a deep learning model called CSMoFN (cancellable SoftmaxOut fusion network). CSMoFN comprises three modules: a feature extraction and fusion module, an enveloping SoftmaxOut transform module, and a multiplicative diagonal compression module.…”

Deep Learning in the Field of Biometric Template Protection: An Overview

“…However, similar schemes have already been proposed for other biometric characteristics as well as multibiometric systems, e.g. in [61], [62]. While the reported results of these very recently proposed methods are promising, a rigorous analysis of potential vulnerabilities is needed.…”

Modified Firefly Optimization with Deep Learning based Multimodal Biometric Verification Model

“…Multimodal biometrics including template protection was not a new subject, several studies were published related to this subject. But DL related multimodal biometric techniques that add template protection persist very scarcely [21][22][23].…”