Using Shannon Entropy as EEG Signal Feature for Fast Person Identification

Dat TRAN, Wanli MA, Phuoc Nguyen

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    18 Citations (Scopus)

    Abstract

    Identification accuracy and speed are important factors in automatic person identification systems. In this paper, we propose a feature extraction method to extract brain wave features from different brain rhythms of electroencephalography (EEG) signal for the purpose of fast, yet accurate person identification. The proposed feature extraction method is based on the fact that EEG signal is complex, non-stationary, and non-linear. With this fact, non-linear analysis like entropy would be more appropriate. Shannon entropy (SE) based EEG features from alpha, beta, and gamma wave bands are extracted and evaluated for person identification. Experimental results show that SE features provide high person identification rates yet with a low feature dimension, thus better performance.
    Original languageEnglish
    Title of host publicationESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning
    EditorsMichel Verleysen
    Place of PublicationBelgium
    PublisherSymposium on Artifical Neural Networks
    Pages413-418
    Number of pages6
    Volume1
    ISBN (Print)9782874190957
    Publication statusPublished - 23 Apr 2014
    EventEuropean Symposium on Artificial Neural Networks: ESANN 2014 - Brugge, Brugge, Belgium
    Duration: 23 Apr 201425 Apr 2014

    Conference

    ConferenceEuropean Symposium on Artificial Neural Networks
    Abbreviated titleESANN 2014
    CountryBelgium
    CityBrugge
    Period23/04/1425/04/14

    Fingerprint

    Electroencephalography
    Entropy
    Feature extraction
    Brain
    Nonlinear analysis
    Identification (control systems)

    Cite this

    TRAN, D., MA, W., & Nguyen, P. (2014). Using Shannon Entropy as EEG Signal Feature for Fast Person Identification. In M. Verleysen (Ed.), ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning (Vol. 1, pp. 413-418). Belgium: Symposium on Artifical Neural Networks.
    TRAN, Dat ; MA, Wanli ; Nguyen, Phuoc. / Using Shannon Entropy as EEG Signal Feature for Fast Person Identification. ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning. editor / Michel Verleysen. Vol. 1 Belgium : Symposium on Artifical Neural Networks, 2014. pp. 413-418
    @inproceedings{5be2b447efdb4c6b8f2ab15d7cc88bc8,
    title = "Using Shannon Entropy as EEG Signal Feature for Fast Person Identification",
    abstract = "Identification accuracy and speed are important factors in automatic person identification systems. In this paper, we propose a feature extraction method to extract brain wave features from different brain rhythms of electroencephalography (EEG) signal for the purpose of fast, yet accurate person identification. The proposed feature extraction method is based on the fact that EEG signal is complex, non-stationary, and non-linear. With this fact, non-linear analysis like entropy would be more appropriate. Shannon entropy (SE) based EEG features from alpha, beta, and gamma wave bands are extracted and evaluated for person identification. Experimental results show that SE features provide high person identification rates yet with a low feature dimension, thus better performance.",
    keywords = "Shannon-Entropy, Person-Identification, EEG",
    author = "Dat TRAN and Wanli MA and Phuoc Nguyen",
    year = "2014",
    month = "4",
    day = "23",
    language = "English",
    isbn = "9782874190957",
    volume = "1",
    pages = "413--418",
    editor = "Michel Verleysen",
    booktitle = "ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning",
    publisher = "Symposium on Artifical Neural Networks",

    }

    TRAN, D, MA, W & Nguyen, P 2014, Using Shannon Entropy as EEG Signal Feature for Fast Person Identification. in M Verleysen (ed.), ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning. vol. 1, Symposium on Artifical Neural Networks, Belgium, pp. 413-418, European Symposium on Artificial Neural Networks, Brugge, Belgium, 23/04/14.

    Using Shannon Entropy as EEG Signal Feature for Fast Person Identification. / TRAN, Dat; MA, Wanli; Nguyen, Phuoc.

    ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning. ed. / Michel Verleysen. Vol. 1 Belgium : Symposium on Artifical Neural Networks, 2014. p. 413-418.

    Research output: A Conference proceeding or a Chapter in BookConference contribution

    TY - GEN

    T1 - Using Shannon Entropy as EEG Signal Feature for Fast Person Identification

    AU - TRAN, Dat

    AU - MA, Wanli

    AU - Nguyen, Phuoc

    PY - 2014/4/23

    Y1 - 2014/4/23

    N2 - Identification accuracy and speed are important factors in automatic person identification systems. In this paper, we propose a feature extraction method to extract brain wave features from different brain rhythms of electroencephalography (EEG) signal for the purpose of fast, yet accurate person identification. The proposed feature extraction method is based on the fact that EEG signal is complex, non-stationary, and non-linear. With this fact, non-linear analysis like entropy would be more appropriate. Shannon entropy (SE) based EEG features from alpha, beta, and gamma wave bands are extracted and evaluated for person identification. Experimental results show that SE features provide high person identification rates yet with a low feature dimension, thus better performance.

    AB - Identification accuracy and speed are important factors in automatic person identification systems. In this paper, we propose a feature extraction method to extract brain wave features from different brain rhythms of electroencephalography (EEG) signal for the purpose of fast, yet accurate person identification. The proposed feature extraction method is based on the fact that EEG signal is complex, non-stationary, and non-linear. With this fact, non-linear analysis like entropy would be more appropriate. Shannon entropy (SE) based EEG features from alpha, beta, and gamma wave bands are extracted and evaluated for person identification. Experimental results show that SE features provide high person identification rates yet with a low feature dimension, thus better performance.

    KW - Shannon-Entropy

    KW - Person-Identification

    KW - EEG

    M3 - Conference contribution

    SN - 9782874190957

    VL - 1

    SP - 413

    EP - 418

    BT - ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning

    A2 - Verleysen, Michel

    PB - Symposium on Artifical Neural Networks

    CY - Belgium

    ER -

    TRAN D, MA W, Nguyen P. Using Shannon Entropy as EEG Signal Feature for Fast Person Identification. In Verleysen M, editor, ESANN 2014 European Symposium on Artifical Neural Networks, Computational Intelligence and Machine Learning. Vol. 1. Belgium: Symposium on Artifical Neural Networks. 2014. p. 413-418