DeepMasterPrints: Generating masterprints for dictionary attacks via latent variable evolution

Philip Bontrager, Aditi Roy, Julian Togelius, Nasir Memon, Arun Ross

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recent research has demonstrated the vulnerability of fingerprint recognition systems to dictionary attacks based on MasterPrints. MasterPrints are real or synthetic fingerprints that can fortuitously match with a large number of fingerprints thereby undermining the security afforded by fingerprint systems. Previous work by Roy et al. generated synthetic MasterPrints at the feature-level. In this work we generate complete image-level MasterPrints known as DeepMasterPrints, whose attack accuracy is found to be much superior than that of previous methods. The proposed method, referred to as Latent Variable Evolution, is based on training a Generative Adversarial Network on a set of real fingerprint images. Stochastic search in the form of the Covariance Matrix Adaptation Evolution Strategy is then used to search for latent input variables to the generator network that can maximize the number of impostor matches as assessed by a fingerprint recognizer. Experiments convey the efficacy of the proposed method in generating DeepMasterPrints. The underlying method is likely to have broad applications in fingerprint security as well as fingerprint synthesis.

Original languageEnglish (US)
Title of host publication2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538671795
DOIs
StatePublished - Apr 24 2019
Event9th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2018 - Redondo Beach, United States
Duration: Oct 22 2018Oct 25 2018

Publication series

Name2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018

Conference

Conference9th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2018
CountryUnited States
CityRedondo Beach
Period10/22/1810/25/18

Fingerprint

Latent Variables
Glossaries
Covariance matrix
Fingerprint
Attack
Experiments
Covariance Matrix Adaptation
Fingerprint Recognition
Stochastic Search
Evolution Strategies
Dictionary
Vulnerability
Efficacy
Maximise
Likely
Generator
Synthesis
Experiment

ASJC Scopus subject areas

  • Statistics and Probability
  • Biomedical Engineering
  • Computer Science Applications

Cite this

Bontrager, P., Roy, A., Togelius, J., Memon, N., & Ross, A. (2019). DeepMasterPrints: Generating masterprints for dictionary attacks via latent variable evolution. In 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018 [8698539] (2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BTAS.2018.8698539

DeepMasterPrints : Generating masterprints for dictionary attacks via latent variable evolution. / Bontrager, Philip; Roy, Aditi; Togelius, Julian; Memon, Nasir; Ross, Arun.

2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. 8698539 (2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bontrager, P, Roy, A, Togelius, J, Memon, N & Ross, A 2019, DeepMasterPrints: Generating masterprints for dictionary attacks via latent variable evolution. in 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018., 8698539, 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018, Institute of Electrical and Electronics Engineers Inc., 9th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2018, Redondo Beach, United States, 10/22/18. https://doi.org/10.1109/BTAS.2018.8698539
Bontrager P, Roy A, Togelius J, Memon N, Ross A. DeepMasterPrints: Generating masterprints for dictionary attacks via latent variable evolution. In 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018. Institute of Electrical and Electronics Engineers Inc. 2019. 8698539. (2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018). https://doi.org/10.1109/BTAS.2018.8698539
Bontrager, Philip ; Roy, Aditi ; Togelius, Julian ; Memon, Nasir ; Ross, Arun. / DeepMasterPrints : Generating masterprints for dictionary attacks via latent variable evolution. 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. (2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018).
@inproceedings{58bcd811b50d4d768e5332712d3704cf,
title = "DeepMasterPrints: Generating masterprints for dictionary attacks via latent variable evolution",
abstract = "Recent research has demonstrated the vulnerability of fingerprint recognition systems to dictionary attacks based on MasterPrints. MasterPrints are real or synthetic fingerprints that can fortuitously match with a large number of fingerprints thereby undermining the security afforded by fingerprint systems. Previous work by Roy et al. generated synthetic MasterPrints at the feature-level. In this work we generate complete image-level MasterPrints known as DeepMasterPrints, whose attack accuracy is found to be much superior than that of previous methods. The proposed method, referred to as Latent Variable Evolution, is based on training a Generative Adversarial Network on a set of real fingerprint images. Stochastic search in the form of the Covariance Matrix Adaptation Evolution Strategy is then used to search for latent input variables to the generator network that can maximize the number of impostor matches as assessed by a fingerprint recognizer. Experiments convey the efficacy of the proposed method in generating DeepMasterPrints. The underlying method is likely to have broad applications in fingerprint security as well as fingerprint synthesis.",
author = "Philip Bontrager and Aditi Roy and Julian Togelius and Nasir Memon and Arun Ross",
year = "2019",
month = "4",
day = "24",
doi = "10.1109/BTAS.2018.8698539",
language = "English (US)",
series = "2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018",

}

TY - GEN

T1 - DeepMasterPrints

T2 - Generating masterprints for dictionary attacks via latent variable evolution

AU - Bontrager, Philip

AU - Roy, Aditi

AU - Togelius, Julian

AU - Memon, Nasir

AU - Ross, Arun

PY - 2019/4/24

Y1 - 2019/4/24

N2 - Recent research has demonstrated the vulnerability of fingerprint recognition systems to dictionary attacks based on MasterPrints. MasterPrints are real or synthetic fingerprints that can fortuitously match with a large number of fingerprints thereby undermining the security afforded by fingerprint systems. Previous work by Roy et al. generated synthetic MasterPrints at the feature-level. In this work we generate complete image-level MasterPrints known as DeepMasterPrints, whose attack accuracy is found to be much superior than that of previous methods. The proposed method, referred to as Latent Variable Evolution, is based on training a Generative Adversarial Network on a set of real fingerprint images. Stochastic search in the form of the Covariance Matrix Adaptation Evolution Strategy is then used to search for latent input variables to the generator network that can maximize the number of impostor matches as assessed by a fingerprint recognizer. Experiments convey the efficacy of the proposed method in generating DeepMasterPrints. The underlying method is likely to have broad applications in fingerprint security as well as fingerprint synthesis.

AB - Recent research has demonstrated the vulnerability of fingerprint recognition systems to dictionary attacks based on MasterPrints. MasterPrints are real or synthetic fingerprints that can fortuitously match with a large number of fingerprints thereby undermining the security afforded by fingerprint systems. Previous work by Roy et al. generated synthetic MasterPrints at the feature-level. In this work we generate complete image-level MasterPrints known as DeepMasterPrints, whose attack accuracy is found to be much superior than that of previous methods. The proposed method, referred to as Latent Variable Evolution, is based on training a Generative Adversarial Network on a set of real fingerprint images. Stochastic search in the form of the Covariance Matrix Adaptation Evolution Strategy is then used to search for latent input variables to the generator network that can maximize the number of impostor matches as assessed by a fingerprint recognizer. Experiments convey the efficacy of the proposed method in generating DeepMasterPrints. The underlying method is likely to have broad applications in fingerprint security as well as fingerprint synthesis.

UR - http://www.scopus.com/inward/record.url?scp=85065397323&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065397323&partnerID=8YFLogxK

U2 - 10.1109/BTAS.2018.8698539

DO - 10.1109/BTAS.2018.8698539

M3 - Conference contribution

AN - SCOPUS:85065397323

T3 - 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018

BT - 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems, BTAS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -