Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy

E.J.C. Kelkboom, K.T.J. de Groot, C. Chen, J. Breebaart, Raymond N.J. Veldhuis

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

4 Citations (Scopus)
42 Downloads (Pure)

Abstract

One of the requirements of a biometric template protection system is that the protected template ideally should not leak any information about the biometric sample or its derivatives. In the literature, several proposed template protection techniques are based on binary vectors. Hence, they require the extraction of a binary representation from the real- valued biometric sample. In this work we focus on the Detection Rate Optimized Bit Allocation (DROBA) quantization scheme that extracts multiple bits per feature component while maximizing the overall detection rate. The allocation strategy has to be stored as auxiliary data for reuse in the verification phase and is considered as public. This implies that the auxiliary data should not leak any information about the extracted binary representation. Experiments in our work show that the original DROBA algorithm, as known in the literature, creates auxiliary data that leaks a significant amount of information. We show how an adversary is able to exploit this information and significantly increase its success rate on obtaining a false accept. Fortunately, the information leakage can be mitigated by restricting the allocation freedom of the DROBA algorithm. We propose a method based on population statistics and empirically illustrate its effectiveness. All the experiments are based on the MCYT fingerprint database using two different texture based feature extraction algorithms.
Original languageUndefined
Title of host publicationIEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.
Place of PublicationPiscataway
PublisherIEEE Computer Society
Pages1-8
Number of pages8
ISBN (Print)978-1-4244-5019-0
DOIs
Publication statusPublished - 28 Sep 2009
Event3rd IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS 2009 - Washington, United States
Duration: 28 Sep 200930 Sep 2009
Conference number: 3

Publication series

Name
PublisherIEEE Computer Society Press

Conference

Conference3rd IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS 2009
Abbreviated titleBTAS
CountryUnited States
CityWashington
Period28/09/0930/09/09

Keywords

  • IR-69008
  • METIS-264264
  • Biometrics
  • EWI-17049
  • Leakage
  • SCS-Safety
  • Template Protection

Cite this

Kelkboom, E. J. C., de Groot, K. T. J., Chen, C., Breebaart, J., & Veldhuis, R. N. J. (2009). Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy. In IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09. (pp. 1-8). [10.1109/BTAS.2009.5339046] Piscataway: IEEE Computer Society. https://doi.org/10.1109/BTAS.2009.5339046
Kelkboom, E.J.C. ; de Groot, K.T.J. ; Chen, C. ; Breebaart, J. ; Veldhuis, Raymond N.J. / Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy. IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.. Piscataway : IEEE Computer Society, 2009. pp. 1-8
@inproceedings{ccd159a254d142a287c9dc7c36f24933,
title = "Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy",
abstract = "One of the requirements of a biometric template protection system is that the protected template ideally should not leak any information about the biometric sample or its derivatives. In the literature, several proposed template protection techniques are based on binary vectors. Hence, they require the extraction of a binary representation from the real- valued biometric sample. In this work we focus on the Detection Rate Optimized Bit Allocation (DROBA) quantization scheme that extracts multiple bits per feature component while maximizing the overall detection rate. The allocation strategy has to be stored as auxiliary data for reuse in the verification phase and is considered as public. This implies that the auxiliary data should not leak any information about the extracted binary representation. Experiments in our work show that the original DROBA algorithm, as known in the literature, creates auxiliary data that leaks a significant amount of information. We show how an adversary is able to exploit this information and significantly increase its success rate on obtaining a false accept. Fortunately, the information leakage can be mitigated by restricting the allocation freedom of the DROBA algorithm. We propose a method based on population statistics and empirically illustrate its effectiveness. All the experiments are based on the MCYT fingerprint database using two different texture based feature extraction algorithms.",
keywords = "IR-69008, METIS-264264, Biometrics, EWI-17049, Leakage, SCS-Safety, Template Protection",
author = "E.J.C. Kelkboom and {de Groot}, K.T.J. and C. Chen and J. Breebaart and Veldhuis, {Raymond N.J.}",
note = "10.1109/BTAS.2009.5339046",
year = "2009",
month = "9",
day = "28",
doi = "10.1109/BTAS.2009.5339046",
language = "Undefined",
isbn = "978-1-4244-5019-0",
publisher = "IEEE Computer Society",
pages = "1--8",
booktitle = "IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.",
address = "United States",

}

Kelkboom, EJC, de Groot, KTJ, Chen, C, Breebaart, J & Veldhuis, RNJ 2009, Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy. in IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.., 10.1109/BTAS.2009.5339046, IEEE Computer Society, Piscataway, pp. 1-8, 3rd IEEE International Conference on Biometrics: Theory, Applications, and Systems, BTAS 2009, Washington, United States, 28/09/09. https://doi.org/10.1109/BTAS.2009.5339046

Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy. / Kelkboom, E.J.C.; de Groot, K.T.J.; Chen, C.; Breebaart, J.; Veldhuis, Raymond N.J.

IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.. Piscataway : IEEE Computer Society, 2009. p. 1-8 10.1109/BTAS.2009.5339046.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy

AU - Kelkboom, E.J.C.

AU - de Groot, K.T.J.

AU - Chen, C.

AU - Breebaart, J.

AU - Veldhuis, Raymond N.J.

N1 - 10.1109/BTAS.2009.5339046

PY - 2009/9/28

Y1 - 2009/9/28

N2 - One of the requirements of a biometric template protection system is that the protected template ideally should not leak any information about the biometric sample or its derivatives. In the literature, several proposed template protection techniques are based on binary vectors. Hence, they require the extraction of a binary representation from the real- valued biometric sample. In this work we focus on the Detection Rate Optimized Bit Allocation (DROBA) quantization scheme that extracts multiple bits per feature component while maximizing the overall detection rate. The allocation strategy has to be stored as auxiliary data for reuse in the verification phase and is considered as public. This implies that the auxiliary data should not leak any information about the extracted binary representation. Experiments in our work show that the original DROBA algorithm, as known in the literature, creates auxiliary data that leaks a significant amount of information. We show how an adversary is able to exploit this information and significantly increase its success rate on obtaining a false accept. Fortunately, the information leakage can be mitigated by restricting the allocation freedom of the DROBA algorithm. We propose a method based on population statistics and empirically illustrate its effectiveness. All the experiments are based on the MCYT fingerprint database using two different texture based feature extraction algorithms.

AB - One of the requirements of a biometric template protection system is that the protected template ideally should not leak any information about the biometric sample or its derivatives. In the literature, several proposed template protection techniques are based on binary vectors. Hence, they require the extraction of a binary representation from the real- valued biometric sample. In this work we focus on the Detection Rate Optimized Bit Allocation (DROBA) quantization scheme that extracts multiple bits per feature component while maximizing the overall detection rate. The allocation strategy has to be stored as auxiliary data for reuse in the verification phase and is considered as public. This implies that the auxiliary data should not leak any information about the extracted binary representation. Experiments in our work show that the original DROBA algorithm, as known in the literature, creates auxiliary data that leaks a significant amount of information. We show how an adversary is able to exploit this information and significantly increase its success rate on obtaining a false accept. Fortunately, the information leakage can be mitigated by restricting the allocation freedom of the DROBA algorithm. We propose a method based on population statistics and empirically illustrate its effectiveness. All the experiments are based on the MCYT fingerprint database using two different texture based feature extraction algorithms.

KW - IR-69008

KW - METIS-264264

KW - Biometrics

KW - EWI-17049

KW - Leakage

KW - SCS-Safety

KW - Template Protection

U2 - 10.1109/BTAS.2009.5339046

DO - 10.1109/BTAS.2009.5339046

M3 - Conference contribution

SN - 978-1-4244-5019-0

SP - 1

EP - 8

BT - IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.

PB - IEEE Computer Society

CY - Piscataway

ER -

Kelkboom EJC, de Groot KTJ, Chen C, Breebaart J, Veldhuis RNJ. Pitfall of the Detection Rate Optimized Bit Allocation within template protection and a remedy. In IEEE 3rd International Conference on Biometrics: Theory, Applications, and Systems, 2009. BTAS '09.. Piscataway: IEEE Computer Society. 2009. p. 1-8. 10.1109/BTAS.2009.5339046 https://doi.org/10.1109/BTAS.2009.5339046