Mimicking human texture classification

B.E. Rogowitz (Editor), Eva M. van Rikxoort, Egon van den Broek, T.N. Pappas (Editor), Theo E. Schouten, S.J. Daly (Editor)

    Research output: Contribution to conferencePaper

    6 Citations (Scopus)
    29 Downloads (Pure)

    Abstract

    In an attempt to mimic human (colorful) texture classification by a clustering algorithm three lines of research have been encountered, in which as test set 180 texture images (both their color and gray-scale equivalent) were drawn from the OuTex and VisTex databases. First, a k-means algorithm was applied with three feature vectors, based on color/gray values, four texture features, and their combination. Second, 18 participants clustered the images using a newly developed card sorting program. The mutual agreement between the participants was 57% and 56% and between the algorithm and the participants it was 47% and 45%, for respectively color and gray-scale texture images. Third, in a benchmark, 30 participants judged the algorithms’ clusters with gray-scale textures as more homogeneous then those with colored textures. However, a high interpersonal variability was present for both the color and the gray-scale clusters. So, despite the promising results, it is questionable whether average human texture classification can be mimicked (if it exists at all).
    Original languageUndefined
    Pages215-226
    Number of pages12
    DOIs
    Publication statusPublished - 18 Mar 2005

    Keywords

    • Texture
    • card sorting
    • k-Means
    • mimic
    • EWI-21116
    • Color
    • HMI-HF: Human Factors
    • IR-59827
    • Human texture perception
    • Clustering
    • HMI-CI: Computational Intelligence

    Cite this

    Rogowitz, B. E. (Ed.), van Rikxoort, E. M., van den Broek, E., Pappas, T. N. (Ed.), Schouten, T. E., & Daly, S. J. (Ed.) (2005). Mimicking human texture classification. 215-226. https://doi.org/10.1117/12.587942
    Rogowitz, B.E. (Editor) ; van Rikxoort, Eva M. ; van den Broek, Egon ; Pappas, T.N. (Editor) ; Schouten, Theo E. ; Daly, S.J. (Editor). / Mimicking human texture classification. 12 p.
    @conference{46f8a4fd903943768aed2bdf2aad57b9,
    title = "Mimicking human texture classification",
    abstract = "In an attempt to mimic human (colorful) texture classification by a clustering algorithm three lines of research have been encountered, in which as test set 180 texture images (both their color and gray-scale equivalent) were drawn from the OuTex and VisTex databases. First, a k-means algorithm was applied with three feature vectors, based on color/gray values, four texture features, and their combination. Second, 18 participants clustered the images using a newly developed card sorting program. The mutual agreement between the participants was 57{\%} and 56{\%} and between the algorithm and the participants it was 47{\%} and 45{\%}, for respectively color and gray-scale texture images. Third, in a benchmark, 30 participants judged the algorithms’ clusters with gray-scale textures as more homogeneous then those with colored textures. However, a high interpersonal variability was present for both the color and the gray-scale clusters. So, despite the promising results, it is questionable whether average human texture classification can be mimicked (if it exists at all).",
    keywords = "Texture, card sorting, k-Means, mimic, EWI-21116, Color, HMI-HF: Human Factors, IR-59827, Human texture perception, Clustering, HMI-CI: Computational Intelligence",
    author = "B.E. Rogowitz and {van Rikxoort}, {Eva M.} and {van den Broek}, Egon and T.N. Pappas and Schouten, {Theo E.} and S.J. Daly",
    year = "2005",
    month = "3",
    day = "18",
    doi = "10.1117/12.587942",
    language = "Undefined",
    pages = "215--226",

    }

    Rogowitz, BE (ed.), van Rikxoort, EM, van den Broek, E, Pappas, TN (ed.), Schouten, TE & Daly, SJ (ed.) 2005, 'Mimicking human texture classification' pp. 215-226. https://doi.org/10.1117/12.587942

    Mimicking human texture classification. / Rogowitz, B.E. (Editor); van Rikxoort, Eva M.; van den Broek, Egon; Pappas, T.N. (Editor); Schouten, Theo E.; Daly, S.J. (Editor).

    2005. 215-226.

    Research output: Contribution to conferencePaper

    TY - CONF

    T1 - Mimicking human texture classification

    AU - van Rikxoort, Eva M.

    AU - van den Broek, Egon

    AU - Schouten, Theo E.

    A2 - Rogowitz, B.E.

    A2 - Pappas, T.N.

    A2 - Daly, S.J.

    PY - 2005/3/18

    Y1 - 2005/3/18

    N2 - In an attempt to mimic human (colorful) texture classification by a clustering algorithm three lines of research have been encountered, in which as test set 180 texture images (both their color and gray-scale equivalent) were drawn from the OuTex and VisTex databases. First, a k-means algorithm was applied with three feature vectors, based on color/gray values, four texture features, and their combination. Second, 18 participants clustered the images using a newly developed card sorting program. The mutual agreement between the participants was 57% and 56% and between the algorithm and the participants it was 47% and 45%, for respectively color and gray-scale texture images. Third, in a benchmark, 30 participants judged the algorithms’ clusters with gray-scale textures as more homogeneous then those with colored textures. However, a high interpersonal variability was present for both the color and the gray-scale clusters. So, despite the promising results, it is questionable whether average human texture classification can be mimicked (if it exists at all).

    AB - In an attempt to mimic human (colorful) texture classification by a clustering algorithm three lines of research have been encountered, in which as test set 180 texture images (both their color and gray-scale equivalent) were drawn from the OuTex and VisTex databases. First, a k-means algorithm was applied with three feature vectors, based on color/gray values, four texture features, and their combination. Second, 18 participants clustered the images using a newly developed card sorting program. The mutual agreement between the participants was 57% and 56% and between the algorithm and the participants it was 47% and 45%, for respectively color and gray-scale texture images. Third, in a benchmark, 30 participants judged the algorithms’ clusters with gray-scale textures as more homogeneous then those with colored textures. However, a high interpersonal variability was present for both the color and the gray-scale clusters. So, despite the promising results, it is questionable whether average human texture classification can be mimicked (if it exists at all).

    KW - Texture

    KW - card sorting

    KW - k-Means

    KW - mimic

    KW - EWI-21116

    KW - Color

    KW - HMI-HF: Human Factors

    KW - IR-59827

    KW - Human texture perception

    KW - Clustering

    KW - HMI-CI: Computational Intelligence

    U2 - 10.1117/12.587942

    DO - 10.1117/12.587942

    M3 - Paper

    SP - 215

    EP - 226

    ER -

    Rogowitz BE, (ed.), van Rikxoort EM, van den Broek E, Pappas TN, (ed.), Schouten TE, Daly SJ, (ed.). Mimicking human texture classification. 2005. https://doi.org/10.1117/12.587942