On the Practicality of Local Ranking-Based Cancelable Iris Recognition

Abstract

Practical cancelable biometrics (CB) schemes should satisfy the requirements of revocability, non-invertibility, and non-linkability without deteriorating the matching accuracy of the underlying biometric recognition system. In order to bridge the gap between theory and practice, it is important to prove that new CB schemes can achieve a balance between the conflicting goals of security and matching accuracy. This paper investigates the security and accuracy trade-off of a recently proposed local ranking-based cancelable biometrics (LRCB) scheme for protecting iris-codes. First, the irreversibility of the LRCB is revisited and a new attack for reversing the LRCB transform is proposed. The proposed attack utilizes the distribution of order statistics for discrete random variables to reverse the protected rank values and obtain a close approximation of the original iris template. This ranking-inversion attack is then utilized to realize authentication, record multiplicity, and correlation attacks against the LRCB transform. Our theoretical analysis shows that the proposed reversibility attack can recover more than 95% of the original iris-code bits and the proposed correlation attack can correctly correlate two templates 100% of the time for the parameter settings that retain the recognition accuracy of the underlying iris recognition system. The validity of the proposed attacks is verified using the same iris dataset adopted by the authors of the LRCB scheme. Experimental results support our theoretical findings and demonstrate that the security properties of irreversibility and non-linkability can only be fulfilled at the expense of significant and impractical degradation of the matching accuracy.INDEX TERMS Cancelable biometrics, iris recognition, local ranking, order statistics, irreversibility, attacks via record multiplicity, linkability attack.

A standard ranking algorithm for robust iris template protection

“…However, challenges arise in meeting specified string length requirements, ranging from 196,608 bits in basic methods to 1,536 bits in the shift technique, potentially compromising irreversibility. According to Ouda [15], the local rank algorithm did not fulfill the irreversibility and unlinkability requirements of BTP Another research from Dwivedi et al [16] suggested a row-wise scheme with decimal encodingbased look-up table mapping (LUT), advocating for decimal encoding and LUT, presenting a deterministic approach but demanding secure LUT generation. Despite ensuring that extracted bits originate from the same input data, the methodology's reliance on a 32,768-bit iris length and the need for 4 iris images per subject pose constraints, echoing accuracy concerns in poor-quality image scenarios.…”

“…This algorithm utilizes user-specific keys to transform the extracted iris code after applying the local rank, generating transformed templates. Notably, according to [15], the proposed algorithm in [14] did not fulfill the irreversibility and non-linkability requirements of BTP. In another indexing method explored in valuable research [16], a row-wise scheme with decimal encoding-based look-up table (LUT) mapping is suggested to transform the iris code into an irreversible representation.…”

Current Multi-factor of Authentication: Approaches, Requirements, Attacks and Challenges

“…[41] RSA Conference in San Francisco, In February 2019, the security expert demonstrated the success of real-time attacks like social engineering schemes, phishing SMS and emails, session hijack, and MITM attacks to intercept the traffic to circumvent the MFA [41] and [42]. In academic research, many attacks are proposed as [43] proposed an attack that used the distribution of order statistics to reverse the protected iris template of the ordinal ranking value of the www.ijacsa.thesai.org original stored iris templates. The proposed reverse-attack successes to recover greater than 95% of the template and can correctly correlate two templates of 100%.…”