Watermarking security incorporating natural scene statistics

Jiangqun Ni, Rongyue Zhang, Chen Fang, Jiwu Huang, Chuntao Wang, Hyong Joong Kim

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

3 Citations (Scopus)

Abstract

Watermarking security has emerged as the domain of extensive research in recent years. This paper presents both information theoretic analysis and practical attack algorithm for spread-spectrum based watermarking security incorporating natural scene statistics (NSS) model. Firstly, the Gaussian scale mixture (GSM) is introduced as the NSS model. The security is quantified by the mutual information between the observed watermarked signals and the secret carriers. The new security measures are then derived based on the GSM model, which allows for more accurate evaluation of watermarking security. Finally, the practical attack algorithm is developed in the framework of variational Bayesian ICA, which is shown to increase the performance and flexibility by allowing incorporation of prior knowledge of host signal. Extensive simulations are carried out to demonstrate the feasibility and effectiveness of the proposed algorithm.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages132-146
Number of pages15
Volume5284 LNCS
DOIs
Publication statusPublished - 2008 Dec 1
Event10th International Workshop on Information Hiding, IH 2008 - Santa Barbara, CA, United States
Duration: 2008 May 192008 May 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5284 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other10th International Workshop on Information Hiding, IH 2008
CountryUnited States
CitySanta Barbara, CA
Period08/5/1908/5/21

Fingerprint

Watermarking
Statistics
Scale Mixture
Gaussian Mixture
Information analysis
Independent component analysis
Attack
Spread Spectrum
Mutual Information
Mixture Model
Prior Knowledge
Flexibility
Evaluation
Model
Demonstrate
Simulation

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Ni, J., Zhang, R., Fang, C., Huang, J., Wang, C., & Kim, H. J. (2008). Watermarking security incorporating natural scene statistics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5284 LNCS, pp. 132-146). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5284 LNCS). https://doi.org/10.1007/978-3-540-88961-8-10

Watermarking security incorporating natural scene statistics. / Ni, Jiangqun; Zhang, Rongyue; Fang, Chen; Huang, Jiwu; Wang, Chuntao; Kim, Hyong Joong.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5284 LNCS 2008. p. 132-146 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5284 LNCS).

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

Ni, J, Zhang, R, Fang, C, Huang, J, Wang, C & Kim, HJ 2008, Watermarking security incorporating natural scene statistics. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 5284 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5284 LNCS, pp. 132-146, 10th International Workshop on Information Hiding, IH 2008, Santa Barbara, CA, United States, 08/5/19. https://doi.org/10.1007/978-3-540-88961-8-10
Ni J, Zhang R, Fang C, Huang J, Wang C, Kim HJ. Watermarking security incorporating natural scene statistics. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5284 LNCS. 2008. p. 132-146. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-540-88961-8-10
Ni, Jiangqun ; Zhang, Rongyue ; Fang, Chen ; Huang, Jiwu ; Wang, Chuntao ; Kim, Hyong Joong. / Watermarking security incorporating natural scene statistics. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5284 LNCS 2008. pp. 132-146 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{6f64595171f346ba894152dd22de0166,
title = "Watermarking security incorporating natural scene statistics",
abstract = "Watermarking security has emerged as the domain of extensive research in recent years. This paper presents both information theoretic analysis and practical attack algorithm for spread-spectrum based watermarking security incorporating natural scene statistics (NSS) model. Firstly, the Gaussian scale mixture (GSM) is introduced as the NSS model. The security is quantified by the mutual information between the observed watermarked signals and the secret carriers. The new security measures are then derived based on the GSM model, which allows for more accurate evaluation of watermarking security. Finally, the practical attack algorithm is developed in the framework of variational Bayesian ICA, which is shown to increase the performance and flexibility by allowing incorporation of prior knowledge of host signal. Extensive simulations are carried out to demonstrate the feasibility and effectiveness of the proposed algorithm.",
author = "Jiangqun Ni and Rongyue Zhang and Chen Fang and Jiwu Huang and Chuntao Wang and Kim, {Hyong Joong}",
year = "2008",
month = "12",
day = "1",
doi = "10.1007/978-3-540-88961-8-10",
language = "English",
isbn = "3540889604",
volume = "5284 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "132--146",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Watermarking security incorporating natural scene statistics

AU - Ni, Jiangqun

AU - Zhang, Rongyue

AU - Fang, Chen

AU - Huang, Jiwu

AU - Wang, Chuntao

AU - Kim, Hyong Joong

PY - 2008/12/1

Y1 - 2008/12/1

N2 - Watermarking security has emerged as the domain of extensive research in recent years. This paper presents both information theoretic analysis and practical attack algorithm for spread-spectrum based watermarking security incorporating natural scene statistics (NSS) model. Firstly, the Gaussian scale mixture (GSM) is introduced as the NSS model. The security is quantified by the mutual information between the observed watermarked signals and the secret carriers. The new security measures are then derived based on the GSM model, which allows for more accurate evaluation of watermarking security. Finally, the practical attack algorithm is developed in the framework of variational Bayesian ICA, which is shown to increase the performance and flexibility by allowing incorporation of prior knowledge of host signal. Extensive simulations are carried out to demonstrate the feasibility and effectiveness of the proposed algorithm.

AB - Watermarking security has emerged as the domain of extensive research in recent years. This paper presents both information theoretic analysis and practical attack algorithm for spread-spectrum based watermarking security incorporating natural scene statistics (NSS) model. Firstly, the Gaussian scale mixture (GSM) is introduced as the NSS model. The security is quantified by the mutual information between the observed watermarked signals and the secret carriers. The new security measures are then derived based on the GSM model, which allows for more accurate evaluation of watermarking security. Finally, the practical attack algorithm is developed in the framework of variational Bayesian ICA, which is shown to increase the performance and flexibility by allowing incorporation of prior knowledge of host signal. Extensive simulations are carried out to demonstrate the feasibility and effectiveness of the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=56749154333&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=56749154333&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-88961-8-10

DO - 10.1007/978-3-540-88961-8-10

M3 - Conference contribution

SN - 3540889604

SN - 9783540889601

VL - 5284 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 132

EP - 146

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -