Content-Based Image Retrieval Benchmarking: Utilizing color categories and color distributions

Egon van den Broek, Peter M.F. Kisters, Louis G. Vuurpijl

    Research output: Contribution to journalArticleAcademicpeer-review

    13 Citations (Scopus)
    71 Downloads (Pure)

    Abstract

    From a human centered perspective three ingredients for Content-Based Image Retrieval (CBIR) were developed. First, with their existence confirmed by experimental data, 11 color categories were utilized for CBIR and used as input for a new color space segmentation technique. The complete HSI color space was divided into 11 segments (or bins), resulting in a unique CBIR 11 color quantization scheme. Second, a new weighted similarity function was introduced. It exploits within bin statistics, describing the distribution of color within a bin. Third, a new CBIR benchmark was successfully used to evaluate both new techniques. Based on the 4050 queries judged by the users, the 11 bin color quantization proved to be useful for CBIR purposes. Moreover, the new weighted similarity function significantly improved retrieval performance, according to the users.
    Original languageUndefined
    Pages (from-to)293-301
    Number of pages9
    JournalJournal of imaging science and technology
    Volume49
    Issue number3
    Publication statusPublished - May 2005

    Keywords

    • HMI-HF: Human Factors
    • HMI-MR: MULTIMEDIA RETRIEVAL
    • color categories
    • weighted similarity function
    • Benchmark
    • Human-Centered
    • IR-78699
    • Content-Based Image Retrieval (CBIR)
    • EWI-20842

    Cite this

    van den Broek, Egon ; Kisters, Peter M.F. ; Vuurpijl, Louis G. / Content-Based Image Retrieval Benchmarking: Utilizing color categories and color distributions. In: Journal of imaging science and technology. 2005 ; Vol. 49, No. 3. pp. 293-301.
    @article{4045716b682d4c509f29dbb4b8d03712,
    title = "Content-Based Image Retrieval Benchmarking: Utilizing color categories and color distributions",
    abstract = "From a human centered perspective three ingredients for Content-Based Image Retrieval (CBIR) were developed. First, with their existence confirmed by experimental data, 11 color categories were utilized for CBIR and used as input for a new color space segmentation technique. The complete HSI color space was divided into 11 segments (or bins), resulting in a unique CBIR 11 color quantization scheme. Second, a new weighted similarity function was introduced. It exploits within bin statistics, describing the distribution of color within a bin. Third, a new CBIR benchmark was successfully used to evaluate both new techniques. Based on the 4050 queries judged by the users, the 11 bin color quantization proved to be useful for CBIR purposes. Moreover, the new weighted similarity function significantly improved retrieval performance, according to the users.",
    keywords = "HMI-HF: Human Factors, HMI-MR: MULTIMEDIA RETRIEVAL, color categories, weighted similarity function, Benchmark, Human-Centered, IR-78699, Content-Based Image Retrieval (CBIR), EWI-20842",
    author = "{van den Broek}, Egon and Kisters, {Peter M.F.} and Vuurpijl, {Louis G.}",
    year = "2005",
    month = "5",
    language = "Undefined",
    volume = "49",
    pages = "293--301",
    journal = "Journal of imaging science and technology",
    issn = "1062-3701",
    publisher = "Society for Imaging Science and Technology",
    number = "3",

    }

    Content-Based Image Retrieval Benchmarking: Utilizing color categories and color distributions. / van den Broek, Egon; Kisters, Peter M.F.; Vuurpijl, Louis G.

    In: Journal of imaging science and technology, Vol. 49, No. 3, 05.2005, p. 293-301.

    Research output: Contribution to journalArticleAcademicpeer-review

    TY - JOUR

    T1 - Content-Based Image Retrieval Benchmarking: Utilizing color categories and color distributions

    AU - van den Broek, Egon

    AU - Kisters, Peter M.F.

    AU - Vuurpijl, Louis G.

    PY - 2005/5

    Y1 - 2005/5

    N2 - From a human centered perspective three ingredients for Content-Based Image Retrieval (CBIR) were developed. First, with their existence confirmed by experimental data, 11 color categories were utilized for CBIR and used as input for a new color space segmentation technique. The complete HSI color space was divided into 11 segments (or bins), resulting in a unique CBIR 11 color quantization scheme. Second, a new weighted similarity function was introduced. It exploits within bin statistics, describing the distribution of color within a bin. Third, a new CBIR benchmark was successfully used to evaluate both new techniques. Based on the 4050 queries judged by the users, the 11 bin color quantization proved to be useful for CBIR purposes. Moreover, the new weighted similarity function significantly improved retrieval performance, according to the users.

    AB - From a human centered perspective three ingredients for Content-Based Image Retrieval (CBIR) were developed. First, with their existence confirmed by experimental data, 11 color categories were utilized for CBIR and used as input for a new color space segmentation technique. The complete HSI color space was divided into 11 segments (or bins), resulting in a unique CBIR 11 color quantization scheme. Second, a new weighted similarity function was introduced. It exploits within bin statistics, describing the distribution of color within a bin. Third, a new CBIR benchmark was successfully used to evaluate both new techniques. Based on the 4050 queries judged by the users, the 11 bin color quantization proved to be useful for CBIR purposes. Moreover, the new weighted similarity function significantly improved retrieval performance, according to the users.

    KW - HMI-HF: Human Factors

    KW - HMI-MR: MULTIMEDIA RETRIEVAL

    KW - color categories

    KW - weighted similarity function

    KW - Benchmark

    KW - Human-Centered

    KW - IR-78699

    KW - Content-Based Image Retrieval (CBIR)

    KW - EWI-20842

    M3 - Article

    VL - 49

    SP - 293

    EP - 301

    JO - Journal of imaging science and technology

    JF - Journal of imaging science and technology

    SN - 1062-3701

    IS - 3

    ER -