A cost-effective decision tree based approach to steganalysis

Liyun Li, Husrev Taha Sencar, Nasir Memon

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

    Abstract

    An important issue concerning real-world deployment of steganalysis systems is the computational cost of acquiring features used in building steganalyzers. Conventional approach to steganalyzer design crucially assumes that all features required for steganalysis have to be computed in advance. However, as the number of features used by typical steganalyzers grow into thousands and timing constraints are imposed on how fast a decision has to be made, this approach becomes impractical. To address this problem, we focus on machine learning aspect of steganalyzer design and introduce a decision tree based approach to steganalysis. The proposed steganalyzer system can minimize the average computational cost for making a steganalysis decision while still maintaining the detection accuracy. To demonstrate the potential of this approach, a series of experiments are performed on well known steganography and steganalysis techniques.

    Original languageEnglish (US)
    Title of host publicationProceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013
    Volume8665
    DOIs
    StatePublished - 2013
    Event2013 Media Watermarking, Security, and Forensics Conference - Burlingame, CA, United States
    Duration: Feb 5 2013Feb 7 2013

    Other

    Other2013 Media Watermarking, Security, and Forensics Conference
    CountryUnited States
    CityBurlingame, CA
    Period2/5/132/7/13

    Fingerprint

    Steganalysis
    Decision trees
    Decision tree
    costs
    Steganography
    Costs
    steganography
    Learning systems
    machine learning
    Computational Cost
    time measurement
    Average Cost
    Experiments
    Timing
    Machine Learning
    Minimise
    Series
    Demonstrate
    Experiment

    Keywords

    • Computational cost effective
    • Decision tree
    • Steganography

    ASJC Scopus subject areas

    • Applied Mathematics
    • Computer Science Applications
    • Electrical and Electronic Engineering
    • Electronic, Optical and Magnetic Materials
    • Condensed Matter Physics

    Cite this

    Li, L., Sencar, H. T., & Memon, N. (2013). A cost-effective decision tree based approach to steganalysis. In Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013 (Vol. 8665). [86650P] https://doi.org/10.1117/12.2008527

    A cost-effective decision tree based approach to steganalysis. / Li, Liyun; Sencar, Husrev Taha; Memon, Nasir.

    Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013. Vol. 8665 2013. 86650P.

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

    Li, L, Sencar, HT & Memon, N 2013, A cost-effective decision tree based approach to steganalysis. in Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013. vol. 8665, 86650P, 2013 Media Watermarking, Security, and Forensics Conference, Burlingame, CA, United States, 2/5/13. https://doi.org/10.1117/12.2008527
    Li L, Sencar HT, Memon N. A cost-effective decision tree based approach to steganalysis. In Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013. Vol. 8665. 2013. 86650P https://doi.org/10.1117/12.2008527
    Li, Liyun ; Sencar, Husrev Taha ; Memon, Nasir. / A cost-effective decision tree based approach to steganalysis. Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013. Vol. 8665 2013.
    @inproceedings{358d69bb2e3744f88a0c91dd18ee8fe8,
    title = "A cost-effective decision tree based approach to steganalysis",
    abstract = "An important issue concerning real-world deployment of steganalysis systems is the computational cost of acquiring features used in building steganalyzers. Conventional approach to steganalyzer design crucially assumes that all features required for steganalysis have to be computed in advance. However, as the number of features used by typical steganalyzers grow into thousands and timing constraints are imposed on how fast a decision has to be made, this approach becomes impractical. To address this problem, we focus on machine learning aspect of steganalyzer design and introduce a decision tree based approach to steganalysis. The proposed steganalyzer system can minimize the average computational cost for making a steganalysis decision while still maintaining the detection accuracy. To demonstrate the potential of this approach, a series of experiments are performed on well known steganography and steganalysis techniques.",
    keywords = "Computational cost effective, Decision tree, Steganography",
    author = "Liyun Li and Sencar, {Husrev Taha} and Nasir Memon",
    year = "2013",
    doi = "10.1117/12.2008527",
    language = "English (US)",
    isbn = "9780819494382",
    volume = "8665",
    booktitle = "Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013",

    }

    TY - GEN

    T1 - A cost-effective decision tree based approach to steganalysis

    AU - Li, Liyun

    AU - Sencar, Husrev Taha

    AU - Memon, Nasir

    PY - 2013

    Y1 - 2013

    N2 - An important issue concerning real-world deployment of steganalysis systems is the computational cost of acquiring features used in building steganalyzers. Conventional approach to steganalyzer design crucially assumes that all features required for steganalysis have to be computed in advance. However, as the number of features used by typical steganalyzers grow into thousands and timing constraints are imposed on how fast a decision has to be made, this approach becomes impractical. To address this problem, we focus on machine learning aspect of steganalyzer design and introduce a decision tree based approach to steganalysis. The proposed steganalyzer system can minimize the average computational cost for making a steganalysis decision while still maintaining the detection accuracy. To demonstrate the potential of this approach, a series of experiments are performed on well known steganography and steganalysis techniques.

    AB - An important issue concerning real-world deployment of steganalysis systems is the computational cost of acquiring features used in building steganalyzers. Conventional approach to steganalyzer design crucially assumes that all features required for steganalysis have to be computed in advance. However, as the number of features used by typical steganalyzers grow into thousands and timing constraints are imposed on how fast a decision has to be made, this approach becomes impractical. To address this problem, we focus on machine learning aspect of steganalyzer design and introduce a decision tree based approach to steganalysis. The proposed steganalyzer system can minimize the average computational cost for making a steganalysis decision while still maintaining the detection accuracy. To demonstrate the potential of this approach, a series of experiments are performed on well known steganography and steganalysis techniques.

    KW - Computational cost effective

    KW - Decision tree

    KW - Steganography

    UR - http://www.scopus.com/inward/record.url?scp=84878432433&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84878432433&partnerID=8YFLogxK

    U2 - 10.1117/12.2008527

    DO - 10.1117/12.2008527

    M3 - Conference contribution

    SN - 9780819494382

    VL - 8665

    BT - Proceedings of SPIE-IS and T Electronic Imaging - Media Watermarking, Security, and Forensics 2013

    ER -