A nonlinear subspace method for pattern recognition using a nonlinear PCA

Ryo Saegusa, Shuji Hashimoto

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

    Abstract

    Linear subspace method based on principal component analysis has been applied for pattern recognition of high-dimensional data. However, the linear subspace method can not represent nonlinear characteristics of the data-distribution efficiently. In order to overcome this problem, some nonlinear subspace methods have been proposed. In this paper, we propose a novel nonlinear subspace method for pattern recognition using multi-layered perceptrons which can hierarchically construct a nonlinear subspace from the data-distribution. We introduce the optimization scheme of the subspace taking into account the classification. The proposed method achieves high classi-fication accuracy by the optimization scheme and provides the finite subspace to avoid the crossover of the subspaces. We examine its effectiveness through some experiments.

    Original languageEnglish
    Title of host publicationProceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006
    Pages45-51
    Number of pages7
    Publication statusPublished - 2006
    Event8th IASTED International Conference on Signal and Image Processing, SIP 2006 and the 10th IASTED International Conference on Internet and Multimedia Systems and Applications, IMSA 2006 - Honolulu, HI
    Duration: 2006 Aug 142006 Aug 16

    Other

    Other8th IASTED International Conference on Signal and Image Processing, SIP 2006 and the 10th IASTED International Conference on Internet and Multimedia Systems and Applications, IMSA 2006
    CityHonolulu, HI
    Period06/8/1406/8/16

    Fingerprint

    Pattern recognition
    Principal component analysis
    Neural networks
    Experiments

    Keywords

    • Eigenspace
    • Feature extraction
    • Neural networks
    • Nonlinear PCA
    • Pattern recognition
    • Subspace method

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition
    • Signal Processing

    Cite this

    Saegusa, R., & Hashimoto, S. (2006). A nonlinear subspace method for pattern recognition using a nonlinear PCA. In Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006 (pp. 45-51)

    A nonlinear subspace method for pattern recognition using a nonlinear PCA. / Saegusa, Ryo; Hashimoto, Shuji.

    Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006. 2006. p. 45-51.

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

    Saegusa, R & Hashimoto, S 2006, A nonlinear subspace method for pattern recognition using a nonlinear PCA. in Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006. pp. 45-51, 8th IASTED International Conference on Signal and Image Processing, SIP 2006 and the 10th IASTED International Conference on Internet and Multimedia Systems and Applications, IMSA 2006, Honolulu, HI, 06/8/14.
    Saegusa R, Hashimoto S. A nonlinear subspace method for pattern recognition using a nonlinear PCA. In Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006. 2006. p. 45-51
    Saegusa, Ryo ; Hashimoto, Shuji. / A nonlinear subspace method for pattern recognition using a nonlinear PCA. Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006. 2006. pp. 45-51
    @inproceedings{93dedd8341c9433aa3558aa75aa1f57e,
    title = "A nonlinear subspace method for pattern recognition using a nonlinear PCA",
    abstract = "Linear subspace method based on principal component analysis has been applied for pattern recognition of high-dimensional data. However, the linear subspace method can not represent nonlinear characteristics of the data-distribution efficiently. In order to overcome this problem, some nonlinear subspace methods have been proposed. In this paper, we propose a novel nonlinear subspace method for pattern recognition using multi-layered perceptrons which can hierarchically construct a nonlinear subspace from the data-distribution. We introduce the optimization scheme of the subspace taking into account the classification. The proposed method achieves high classi-fication accuracy by the optimization scheme and provides the finite subspace to avoid the crossover of the subspaces. We examine its effectiveness through some experiments.",
    keywords = "Eigenspace, Feature extraction, Neural networks, Nonlinear PCA, Pattern recognition, Subspace method",
    author = "Ryo Saegusa and Shuji Hashimoto",
    year = "2006",
    language = "English",
    isbn = "0889865833",
    pages = "45--51",
    booktitle = "Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006",

    }

    TY - GEN

    T1 - A nonlinear subspace method for pattern recognition using a nonlinear PCA

    AU - Saegusa, Ryo

    AU - Hashimoto, Shuji

    PY - 2006

    Y1 - 2006

    N2 - Linear subspace method based on principal component analysis has been applied for pattern recognition of high-dimensional data. However, the linear subspace method can not represent nonlinear characteristics of the data-distribution efficiently. In order to overcome this problem, some nonlinear subspace methods have been proposed. In this paper, we propose a novel nonlinear subspace method for pattern recognition using multi-layered perceptrons which can hierarchically construct a nonlinear subspace from the data-distribution. We introduce the optimization scheme of the subspace taking into account the classification. The proposed method achieves high classi-fication accuracy by the optimization scheme and provides the finite subspace to avoid the crossover of the subspaces. We examine its effectiveness through some experiments.

    AB - Linear subspace method based on principal component analysis has been applied for pattern recognition of high-dimensional data. However, the linear subspace method can not represent nonlinear characteristics of the data-distribution efficiently. In order to overcome this problem, some nonlinear subspace methods have been proposed. In this paper, we propose a novel nonlinear subspace method for pattern recognition using multi-layered perceptrons which can hierarchically construct a nonlinear subspace from the data-distribution. We introduce the optimization scheme of the subspace taking into account the classification. The proposed method achieves high classi-fication accuracy by the optimization scheme and provides the finite subspace to avoid the crossover of the subspaces. We examine its effectiveness through some experiments.

    KW - Eigenspace

    KW - Feature extraction

    KW - Neural networks

    KW - Nonlinear PCA

    KW - Pattern recognition

    KW - Subspace method

    UR - http://www.scopus.com/inward/record.url?scp=56549106275&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=56549106275&partnerID=8YFLogxK

    M3 - Conference contribution

    AN - SCOPUS:56549106275

    SN - 0889865833

    SN - 9780889865839

    SP - 45

    EP - 51

    BT - Proceedings of the 8th IASTED International Conference on Signal and Image Processing, SIP 2006

    ER -