An ensemble of classifiers approach to steganalysis

S. Bayram, A. E. Dirik, H. T. Sencar, N. Memon

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

    Abstract

    Most work on steganalysis, except a few exceptions, have primarily focused on providing features with high discrimination power without giving due consideration to issues concerning practical deployment of steganalysis methods. In this work, we focus on machine learning aspect of steganalyzer design and utilize a hierarchical ensemble of classifiers based approach to tackle two main issues. Firstly, proposed approach provides a workable and systematic procedure to incorporate several steganalyzers together in a composite steganalyzer to improve detection performance in a scalable and cost-effective manner. Secondly, since the approach can be readily extended to multi-class classification it can also be used to infer the steganographic technique deployed in generation of a stego-object. We provide results to demonstrate the potential of the proposed approach.

    Original languageEnglish (US)
    Title of host publicationProceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010
    Pages4376-4379
    Number of pages4
    DOIs
    StatePublished - 2010
    Event2010 20th International Conference on Pattern Recognition, ICPR 2010 - Istanbul, Turkey
    Duration: Aug 23 2010Aug 26 2010

    Other

    Other2010 20th International Conference on Pattern Recognition, ICPR 2010
    CountryTurkey
    CityIstanbul
    Period8/23/108/26/10

    Fingerprint

    Learning systems
    Classifiers
    Composite materials
    Costs

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition

    Cite this

    Bayram, S., Dirik, A. E., Sencar, H. T., & Memon, N. (2010). An ensemble of classifiers approach to steganalysis. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010 (pp. 4376-4379). [5597874] https://doi.org/10.1109/ICPR.2010.1064

    An ensemble of classifiers approach to steganalysis. / Bayram, S.; Dirik, A. E.; Sencar, H. T.; Memon, N.

    Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 4376-4379 5597874.

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

    Bayram, S, Dirik, AE, Sencar, HT & Memon, N 2010, An ensemble of classifiers approach to steganalysis. in Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010., 5597874, pp. 4376-4379, 2010 20th International Conference on Pattern Recognition, ICPR 2010, Istanbul, Turkey, 8/23/10. https://doi.org/10.1109/ICPR.2010.1064
    Bayram S, Dirik AE, Sencar HT, Memon N. An ensemble of classifiers approach to steganalysis. In Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. p. 4376-4379. 5597874 https://doi.org/10.1109/ICPR.2010.1064
    Bayram, S. ; Dirik, A. E. ; Sencar, H. T. ; Memon, N. / An ensemble of classifiers approach to steganalysis. Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010. 2010. pp. 4376-4379
    @inproceedings{03814860a74f497399d63675b614248e,
    title = "An ensemble of classifiers approach to steganalysis",
    abstract = "Most work on steganalysis, except a few exceptions, have primarily focused on providing features with high discrimination power without giving due consideration to issues concerning practical deployment of steganalysis methods. In this work, we focus on machine learning aspect of steganalyzer design and utilize a hierarchical ensemble of classifiers based approach to tackle two main issues. Firstly, proposed approach provides a workable and systematic procedure to incorporate several steganalyzers together in a composite steganalyzer to improve detection performance in a scalable and cost-effective manner. Secondly, since the approach can be readily extended to multi-class classification it can also be used to infer the steganographic technique deployed in generation of a stego-object. We provide results to demonstrate the potential of the proposed approach.",
    author = "S. Bayram and Dirik, {A. E.} and Sencar, {H. T.} and N. Memon",
    year = "2010",
    doi = "10.1109/ICPR.2010.1064",
    language = "English (US)",
    isbn = "9780769541099",
    pages = "4376--4379",
    booktitle = "Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010",

    }

    TY - GEN

    T1 - An ensemble of classifiers approach to steganalysis

    AU - Bayram, S.

    AU - Dirik, A. E.

    AU - Sencar, H. T.

    AU - Memon, N.

    PY - 2010

    Y1 - 2010

    N2 - Most work on steganalysis, except a few exceptions, have primarily focused on providing features with high discrimination power without giving due consideration to issues concerning practical deployment of steganalysis methods. In this work, we focus on machine learning aspect of steganalyzer design and utilize a hierarchical ensemble of classifiers based approach to tackle two main issues. Firstly, proposed approach provides a workable and systematic procedure to incorporate several steganalyzers together in a composite steganalyzer to improve detection performance in a scalable and cost-effective manner. Secondly, since the approach can be readily extended to multi-class classification it can also be used to infer the steganographic technique deployed in generation of a stego-object. We provide results to demonstrate the potential of the proposed approach.

    AB - Most work on steganalysis, except a few exceptions, have primarily focused on providing features with high discrimination power without giving due consideration to issues concerning practical deployment of steganalysis methods. In this work, we focus on machine learning aspect of steganalyzer design and utilize a hierarchical ensemble of classifiers based approach to tackle two main issues. Firstly, proposed approach provides a workable and systematic procedure to incorporate several steganalyzers together in a composite steganalyzer to improve detection performance in a scalable and cost-effective manner. Secondly, since the approach can be readily extended to multi-class classification it can also be used to infer the steganographic technique deployed in generation of a stego-object. We provide results to demonstrate the potential of the proposed approach.

    UR - http://www.scopus.com/inward/record.url?scp=78149474833&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=78149474833&partnerID=8YFLogxK

    U2 - 10.1109/ICPR.2010.1064

    DO - 10.1109/ICPR.2010.1064

    M3 - Conference contribution

    SN - 9780769541099

    SP - 4376

    EP - 4379

    BT - Proceedings - 2010 20th International Conference on Pattern Recognition, ICPR 2010

    ER -