Providing effective protection for iris data is of great importance for the application and popularization of iris biometrics. For the high accuracy but low irreversibility local ranking-based iris template protection and its vulnerablility to ranking inversion attack, two improved schemes based on ordinal value fusion strategy are proposed. The original iris data are first XORed with l different application-specific parameters and ranked to obtain the corresponding ordinal values. After the fusion of the ordinal values, the two improvement schemes further enhance the irreversibility by re-ranking or modulo operations, respectively. The experimental results show that the first improved scheme can achieve an accuracy reduction of 2% or less and an irreversibility index improvement of more than 15% compared with the local ranking scheme, and the second improved scheme can achieve an irreversibility index improvement of more than 30% and an accuracy reduction of 10% or less.
Zhao等[5]提出了一种基于局部排序的可撤销生物识别方案(local ranking based cancelable biometric,LRCB),该方案是近几年比较有代表性的虹膜模板保护方案之一,作者指出该方案能够在满足三大安全性要求的前提下达到与开源虹膜识别系统在原始虹膜数据集上相当的识别精度。Ouda等[6]提出了针对于LRCB方案的顺序逆向攻击模型,实验结果表明,在LRCB方案中精度较高的参数设置下,通过顺序逆向攻击能恢复出大部分原始虹膜数据,说明该方案的不可逆性有待进一步提高。
本文针对LRCB方案容易遭受顺序逆向攻击的问题,提出了基于融合重排的可撤销生物识别(fusion re-ranking based cancelable biometric,FRCB)和基于融合取模的可撤销生物识别(fusion modulo based cancelable biometric,FMCB)。改进方案将LRCB作为其中的部分模块,改进的核心思想在于通过对局部排序的结果进行融合使得攻击者无法准确得到融合之前的顺序值,从而无法发起顺序逆向攻击,并且通过重新排序或取模操作进一步提升不可逆性。
JAINA K, FLYNNP, ROSSA A. Handbook of Biometrics [M]. Berlin: Springer, 2007. DOI: 10.1007/978-0-387-71041-9 .
[2]
RATHGEBC, UHL A. A survey on biometric cryptosystems and cancelable biometrics[J].EURASIP Journal on Information Security, 2011, 2011(1): 1-25. DOI: 10.1186/1687-417X-2011-3 .
[3]
RATHAN K, CONNELLJ H, BOLLER M. Enhancing security and privacy in biometrics-based authentication systems[J]. IBM Systems Journal, 2001,40(3): 614-634. DOI: 10.1147/sj.403.0614 .
[4]
ISO. Information Technology-Security Techniques-Biometric Information Protection[EB/OL]. [2022-05-30]. DOI: 10.3403/30136313 .
[5]
ZHAOD D, FANGS, XIANGJ W, et al. Iris template protection based on local ranking[J]. Security and Communication Networks, 2018, 2018: 1-9. DOI: 10.1155/2018/4519548 .
[6]
OUDAO. On the practicality of local ranking-based cancelable iris recognition[J]. IEEE Access, 2021, 9: 86392-86403. DOI: 10.1109/ACCESS.2021.3089078 .
[7]
RATHGEBC, BREITINGERF, BUSCHC. Alignment-free cancelable iris biometric templates based on adaptive bloom filters[C]//2013 International Conference on Biometrics (ICB). New York: IEEE Press, 2013: 1-8. DOI: 10.1109/ICB.2013.6612976 .
[8]
JINZ, HWANGJ Y, LAIY L, et al. Ranking-based locality sensitive hashing-enabled cancelable biometrics: Index-of-max hashing[J]. IEEE Transactions on Information Forensics and Security, 2018, 13(2): 393-407. DOI: 10.1109/TIFS.2017.2753172 .
[9]
LAIY L, JINZ, TEOHA B J, et al. Cancellable iris template generation based on indexing-first-one Hashing[J]. Pattern Recognition, 2017, 64: 105-117. DOI: 10.1016/j.patcog.2016.10.035 .
[10]
SADHYAD, RAMANB. Generation of cancelable iris templates via randomized bit sampling[J]. IEEE Transactions on Information Forensics and Security, 2019, 14(11): 2972-2986. DOI: 10.1109/TIFS.2019.2907014 .
[11]
TAREKM, HAMOUDAE, ABOHAMAMAA S. Multi-instance cancellable biometrics schemes based on generative adversarial network[J]. Applied Intelligence, 2022, 52(1): 501-513. DOI: 10.1007/s10489-021-02401-7 .
[12]
TAREKM, HAMOUDAE, EL-METWALLYS. Unimodal-Bio-GAN: Keyless biometric salting scheme based on generative adversarial network[J]. IET Biometrics, 2021, 10(6): 654-663. DOI: 10.1049/bme2.12034 .
[13]
TAREKM, OUDAO, HAMZAT. Robust cancellable biometrics scheme based on neural networks[J]. IET Biometrics, 2016, 5(3): 220-228. DOI: 10.1049/iet-bmt.2015.0045 .
[14]
HERMANSJ, MENNINKB, PEETERSR. When a Bloom filter is a Doom filter: Security assessment of a novel iris biometric template protection system[C]//2014 International Conference of the Biometrics Special Interest Group (BIOSIG). New York: IEEE Press, 2015: 1-6.
[15]
DWIVEDIR, DEYS. Cancelable iris template generation using look-up table mapping[C]//2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN). New York: IEEE Press, 2015: 785-790. DOI: 10.1109/SPIN.2015.7095296 .
[16]
ZHAOD D, LUOW J, LIUR, et al. Negative iris recognition[J]. IEEE Transactions on Dependable and Secure Computing, 2018, 15(1): 112-125. DOI: 10.1109/TDSC.2015.2507133 .
[17]
OUDAO, CHAOUIS, TSUMURAN. Security evaluation of negative iris recognition[J]. IEICE Transactions on Information and Systems, 2020, E103.D(5): 1144-1152. DOI: 10.1587/transinf.2019edp7276 .
[18]
ZHAOD D, ZHOUX Y, XIANGJ W, et al. NDBIris with better unlinkability[C]//2020 IEEE Symposium Series on Computational Intelligence (SSCI). New York: IEEE Press, 2021: 2948-2956. DOI: 10.1109/SSCI47803.2020.9308596 .
[19]
YANGW C, WANGS, SHAHZADM, et al. A cancelable biometric authentication system based on feature-adaptive random projection[J]. Journal of Information Security and Applications, 2021, 58: 102704. DOI: 10.1016/j.jisa.2020.102704 .
[20]
EVANSD L, LEEMISL M, DREWJ H. The distribution of order statistics for discrete random variables with applications to bootstrapping[J]. INFORMS Journal on Computing, 2006, 18(1): 19-30. DOI: 10.1287/ijoc.1040.0105 .
OTHMANN, DORIZZIB, GARCIA-SALICETTIS. OSIRIS: An open source iris recognition software[J]. Pattern Recognition Letters, 2016, 82(2):124-131. DOI: 10.1016/j.patrec.2015.09.002 .
[23]
LIC, HUJ K. Attacks via record multiplicity on cancelable biometrics templates[J]. Concurrency and Computation: Practice & Experience, 2014, 26(8): 1593-1605. DOI: 10.1002/cpe.3042 .
[24]
GOMEZ-BARREROM, GALBALLYJ, RATHGEBC, et al. General framework to evaluate unlinkability in biometric template protection systems[J]. IEEE Transactions on Information Forensics and Security, 2018, 13(6): 1406-1420. DOI: 10.1109/TIFS.2017.2788000 .