Abstract:The rapidly evolving field of biometric recognition and verification technologies has seen substantial progress recently. However, the storage of biometric data in cloud environments presents significant security challenges, including data theft and unauthorized modifications. Cancelable biometrics emerge as a strategic solution to these issues. This study examines the efficacy of diverse cancelable biometric methods, classified into transformation, encryption, and hybrid approaches. We introduce a novel cancelable biometric system that combines a hybrid optical asymmetric phase truncated Fourier transform (PTFT) with a symmetric cipher scheme, enhanced by a two-level block-based shuffling technique. This innovation improves security through a streamlined shuffling process that divides the PTFT output into four sections for shuffling and an exclusive OR operations with a unique key, enhancing security and computational efficiency. Our empirical evaluation on three distinct datasets (Faces, Iris, and Fingerprints) from FERET, UPOL and CASIA, and FVC2002 DB1 and DB2, respectively, demonstrates superior performance of our system compared to traditional PTFT-based systems. Notable performance metrics include an Area under the receiver operating characteristic curve of 0.9998. Significant improvements were also observed in false acceptance and false rejection ratios, along with more uniform histogram distributions, thereby enhancing randomness in the resultant deformed patterns. From a security standpoint, our system exhibits a high entropy of 7.95 and achieves near-optimal results in the number of pixels change rate and unified average changing intensity, with values of 98.9115 and 23.9755%, indicating robust resilience against potential differential attacks. The research findings, supported by detailed tables and graphical illustrations, underscore the additional security benefits offered by the two-level encryption mechanism.

Cancellable biometrics based on the index-of-maximum hashing with random sparse binary encoding

Biometrics-based identifiers for digital identity management

Encrypt with Your Mind: Reliable and Revocable Brain Biometrics Via Multidimensional Gaussian Fitted Bit Allocation.

Random permutation‐based linear regression for cancelable biometrics

A cancelable biometric authentication system based on feature-adaptive random projection

On the Risk of Cancelable Biometrics

An efficient multi-biometric cancellable biometric scheme based on deep fusion and deep dream

BioCanCrypto: An LDPC Coded Bio-Cryptosystem on Fingerprint Cancellable Template

Biometric template protection based on a cancelable convolutional neural network over iris and fingerprint

Brain Password: A Secure and Truly Cancelable Brain Biometrics for Smart Headwear.

BioDeepHash: Mapping Biometrics into a Stable Code

Cancellable Template Design for Privacy-Preserving EEG Biometric Authentication Systems

A non-invertible cancelable fingerprint template generation based on ridge feature transformation

Advanced biometric system security: a novel hybrid approach integrating optical asymmetric PTFT and symmetric cipher techniques

Efficient cancelable authentication system based on DRPE and adaptive filter

A cancelable multi-biometric system based on the feature-level fusion of fingerprint and finger vein

A linear convolution-based cancelable fingerprint biometric authentication system

Privacy-preserving biometrics using matrix random low-rank approximation approach

Cancelable Iris template for secure authentication based on random projection and double random phase encoding

Reusable Set Constructions Using Randomized Dissolvent Templates for Biometric Security

Authentication Attacks on Projection-based Cancelable Biometric Schemes