Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
184 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Ranking Based Locality Sensitive Hashing Enabled Cancelable Biometrics: Index-of-Max Hashing (1703.05455v2)

Published 16 Mar 2017 in cs.CV

Abstract: In this paper, we propose a ranking based locality sensitive hashing inspired two-factor cancelable biometrics, dubbed "Index-of-Max" (IoM) hashing for biometric template protection. With externally generated random parameters, IoM hashing transforms a real-valued biometric feature vector into discrete index (max ranked) hashed code. We demonstrate two realizations from IoM hashing notion, namely Gaussian Random Projection based and Uniformly Random Permutation based hashing schemes. The discrete indices representation nature of IoM hashed codes enjoy serveral merits. Firstly, IoM hashing empowers strong concealment to the biometric information. This contributes to the solid ground of non-invertibility guarantee. Secondly, IoM hashing is insensitive to the features magnitude, hence is more robust against biometric features variation. Thirdly, the magnitude-independence trait of IoM hashing makes the hash codes being scale-invariant, which is critical for matching and feature alignment. The experimental results demonstrate favorable accuracy performance on benchmark FVC2002 and FVC2004 fingerprint databases. The analyses justify its resilience to the existing and newly introduced security and privacy attacks as well as satisfy the revocability and unlinkability criteria of cancelable biometrics.

Citations (164)

Summary

  • The paper introduces Index-of-Max (IoM) hashing, a ranking-based LSH technique transforming biometric features into secure, discrete codes for cancelable biometrics.
  • Experimental results on benchmark datasets demonstrate favorable accuracy and robustness against various attacks, including brute-force and ARM, while maintaining key cancelable biometrics criteria.
  • IoM hashing offers practical implications for biometric cybersecurity, potentially integrating with primitives like Fuzzy Vaults and applying to diverse identity authentication scenarios.

Ranking Based Locality Sensitive Hashing Enabled Cancelable Biometrics: Index-of-Max Hashing

The paper introduces a novel approach for biometric template protection, leveraging a two-factor cancelable biometrics strategy called "Index-of-Max" (IoM) hashing. The primary aim is to address concerns related to the security and privacy of biometric templates, particularly when these templates are prone to theft or compromise. IoM hashing employs externally generated random parameters to convert real-valued biometric feature vectors into discrete index-based hashed codes, which are derived from ranking-based locality sensitive hashing (LSH).

The authors outline two specific implementations of IoM hashing: Gaussian Random Projection-based hashing and Uniformly Random Permutation-based hashing. These approaches are designed to provide robust concealment of biometric data, support template non-invertibility, and maintain performance in scenarios involving biometric feature variations. Additionally, IoM hashing ensures scale invariance, which is vital during the matching and feature alignment processes.

Experimental results using benchmark fingerprint databases like FVC2002 and FVC2004 reveal favorable accuracy performance. The methodology meets essential criteria for cancelable biometrics, including non-invertibility, revocability, unlinkability, and performance preservation after transformation. The authors also conduct comprehensive analyses of the security and privacy parameters of IoM hashing, demonstrating its resilience against established and emergent security attacks, such as brute-force, Attack via Record Multiplicity (ARM), false accept, and birthday attacks.

Key Features and Results

  • Biometric Feature Protection: IoM hashing uses ranking-based LSH to hash biometric feature vectors into a discrete index form, ensuring irreversibility and robustness against inversion attacks.
  • Template Revocability: If a biometric template is compromised, IoM hashing allows for easy revocation and regeneration, ensuring the security and privacy of biometric data.
  • Experimental Findings: Favorable accuracy results were recorded on benchmark datasets, with IoM hashing demonstrating strong concealment and resistance against various attack models.
  • Security Analysis: The paper highlights the robustness of IoM hashing against challenging security threats, thereby building confidence in its revocability, unlinkability, and resistance to common biometric attacks.

Implications and Future Work

The IoM hashing scheme presents several practical implications for the field of cybersecurity within biometrics. Its capability to effectively protect biometric templates from unauthorized access while maintaining high accuracy levels creates beneficial avenues for application in systems where data security is paramount.

Furthermore, IoM hashing might form the foundation for future research in integrating it with biometric cryptosystem primitives, like Fuzzy Vault and Fuzzy Commitment, enhancing privacy-preserving capabilities. Exploration into its utility in varied identity authentication scenarios and adaptive application to multi-modal biometric systems also offers promising prospects.

In conclusion, while the work done here reflects considerable advancement in biometric template security, continued refinement of IoM hashing - particularly in terms of complexity and application scope - is essential to maximize its benefits in real-world implementations.