Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models

Marcel Schäfer; Sebastian Mair; Waldemar Berchtold; Martin Steinebach

doi:10.1145/2756601.2756611

Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models

Marcel Schäfer
, Sebastian Mair
, Waldemar Berchtold
, Martin Steinebach

Professur für Wirtschaftsinformatik, insbesondere Machine Learning

Fraunhofer-Institut für Sichere Informationstechnologie

Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › Begutachtung

1 Zitat (Scopus)

Abstract

Collusion attacks on watermarked media copies are commonly countered by probabilistically generated fingerprinting codes and appropriate tracing algorithms. The latter calculates accusation scores representing the suspiciousness of the fingerprints. In a 'detect many' scenario a threshold decides which scores are associated to the colluders. This work proposes a universal method to calculate thresholds for different decoders solely with knowledge of the accusation scores from the actual attack. Applying mixture models on the scores, the threshold is set up satisfying the selected error probabilities. It is independent from the fingerprint generation and can be applied at any decoder. Also no knowledge about the number of attackers or their strategy is needed.

Originalsprache	Englisch
Titel	Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security
Seitenumfang	6
Herausgeber (Verlag)	Association for Computing Machinery, Inc
Erscheinungsdatum	17.06.2015
Seiten	109-114
ISBN (Print)	978-1-4503-3587-4
ISBN (elektronisch)	978-1-4503-3587-4
DOIs	https://doi.org/10.1145/2756601.2756611
Publikationsstatus	Erschienen - 17.06.2015
Veranstaltung	ACM Workshop on Information Hiding and Multimedia Security - 2015 - Portland, USA / Vereinigte Staaten Dauer: 17.06.2015 → 19.06.2015 Konferenznummer: 3 http://sigmm.org/events/acm-workshop-information-hiding-and-multimedia-security-2015

Fachgebiete und Schlagwörter

Informatik

Zugriff auf Dokument

10.1145/2756601.2756611

Dieses zitieren

Schäfer, M., Mair, S., Berchtold, W., & Steinebach, M. (2015). Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models. in Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security (S. 109-114). (IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security). Association for Computing Machinery, Inc. https://doi.org/10.1145/2756601.2756611

Schäfer, Marcel ; Mair, Sebastian ; Berchtold, Waldemar et al. / Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models. Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security. Association for Computing Machinery, Inc, 2015. S. 109-114 (IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security).

@inbook{6077db80f56f4ea998d3cdd4eb23de67,

title = "Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models",

abstract = "Collusion attacks on watermarked media copies are commonly countered by probabilistically generated fingerprinting codes and appropriate tracing algorithms. The latter calculates accusation scores representing the suspiciousness of the fingerprints. In a 'detect many' scenario a threshold decides which scores are associated to the colluders. This work proposes a universal method to calculate thresholds for different decoders solely with knowledge of the accusation scores from the actual attack. Applying mixture models on the scores, the threshold is set up satisfying the selected error probabilities. It is independent from the fingerprint generation and can be applied at any decoder. Also no knowledge about the number of attackers or their strategy is needed.",

keywords = "Informatics, Collusion attacks, Digital watermarking, Fingerprinting codes, Mixture models, Traitor tracing",

author = "Marcel Sch{\"a}fer and Sebastian Mair and Waldemar Berchtold and Martin Steinebach",

year = "2015",

month = jun,

day = "17",

doi = "10.1145/2756601.2756611",

language = "English",

isbn = "978-1-4503-3587-4",

series = "IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security",

publisher = "Association for Computing Machinery, Inc",

pages = "109--114",

booktitle = "Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security",

address = "United States",

note = "ACM Workshop on Information Hiding and Multimedia Security - 2015, ACM Workshop 2015 ; Conference date: 17-06-2015 Through 19-06-2015",

url = "http://sigmm.org/events/acm-workshop-information-hiding-and-multimedia-security-2015",

}

Schäfer, M, Mair, S, Berchtold, W & Steinebach, M 2015, Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models. in Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security. IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security, Association for Computing Machinery, Inc, S. 109-114, ACM Workshop on Information Hiding and Multimedia Security - 2015, Portland, USA / Vereinigte Staaten, 17.06.15. https://doi.org/10.1145/2756601.2756611

Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models. / Schäfer, Marcel; Mair, Sebastian; Berchtold, Waldemar et al.
Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security. Association for Computing Machinery, Inc, 2015. S. 109-114 (IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security).

Publikation: Beiträge in Sammelwerken › Aufsätze in Konferenzbänden › Forschung › Begutachtung

TY - CHAP

T1 - Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models

AU - Schäfer, Marcel

AU - Mair, Sebastian

AU - Berchtold, Waldemar

AU - Steinebach, Martin

N1 - Conference code: 3

PY - 2015/6/17

Y1 - 2015/6/17

N2 - Collusion attacks on watermarked media copies are commonly countered by probabilistically generated fingerprinting codes and appropriate tracing algorithms. The latter calculates accusation scores representing the suspiciousness of the fingerprints. In a 'detect many' scenario a threshold decides which scores are associated to the colluders. This work proposes a universal method to calculate thresholds for different decoders solely with knowledge of the accusation scores from the actual attack. Applying mixture models on the scores, the threshold is set up satisfying the selected error probabilities. It is independent from the fingerprint generation and can be applied at any decoder. Also no knowledge about the number of attackers or their strategy is needed.

AB - Collusion attacks on watermarked media copies are commonly countered by probabilistically generated fingerprinting codes and appropriate tracing algorithms. The latter calculates accusation scores representing the suspiciousness of the fingerprints. In a 'detect many' scenario a threshold decides which scores are associated to the colluders. This work proposes a universal method to calculate thresholds for different decoders solely with knowledge of the accusation scores from the actual attack. Applying mixture models on the scores, the threshold is set up satisfying the selected error probabilities. It is independent from the fingerprint generation and can be applied at any decoder. Also no knowledge about the number of attackers or their strategy is needed.

KW - Informatics

KW - Collusion attacks

KW - Digital watermarking

KW - Fingerprinting codes

KW - Mixture models

KW - Traitor tracing

UR - https://www.scopus.com/pages/publications/84962863575

U2 - 10.1145/2756601.2756611

DO - 10.1145/2756601.2756611

M3 - Article in conference proceedings

SN - 978-1-4503-3587-4

T3 - IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security

SP - 109

EP - 114

BT - Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security

PB - Association for Computing Machinery, Inc

T2 - ACM Workshop on Information Hiding and Multimedia Security - 2015

Y2 - 17 June 2015 through 19 June 2015

ER -

Schäfer M, Mair S, Berchtold W, Steinebach M. Universal Threshold Calculation for Fingerprinting Decoders using Mixture Models. in Proceedings of the 3rd ACM Workshop on Information Hiding and Multimedia Security. Association for Computing Machinery, Inc. 2015. S. 109-114. (IH and MMSec 2015 - Proceedings of the 2015 ACM Workshop on Information Hiding and Multimedia Security). doi: 10.1145/2756601.2756611