Composite Description Based on Salient Contours and Color Information for CBIR Tasks

C. Iakovidou, N. Anagnostopoulos, M. Lux, K. Christodoulou, Y. Boutalis, S. A. Chatzichristofis

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    This paper introduces a novel image descriptor for content-based image retrieval tasks that integrates contour and color information into a compact vector. Loosely inspired by the human visual system and its mechanisms in efficiently identifying visual saliency, operations are performed on a fixed lattice of discrete positions by a set of edge detecting kernels that calculate region derivatives at different scales and orientation. The description method utilizes a weighted edge histogram where bins are populated on the premise of whether the regions contain edges belonging to the salient contours, while the discriminative power is further enhanced by integrating regional quantized color information. The proposed technique is both efficient and adaptive to the specifics of each depiction, while it does not need any training data to adjust parameters. An experimental evaluation conducted on seven benchmarking datasets against 13 well known global descriptors along with SIFT, SURF implementations (both in VLAD and BOVW), highlight the effectiveness and efficiency of the proposed descriptor.

    Original languageEnglish
    Article number8626513
    Pages (from-to)3115-3129
    Number of pages15
    JournalIEEE Transactions on Image Processing
    Volume28
    Issue number6
    DOIs
    Publication statusPublished - 1 Jun 2019

    Fingerprint

    Color
    Composite materials
    Bins
    Image retrieval
    Benchmarking
    Derivatives

    Keywords

    • global features
    • Image retrieval
    • multimedia retrieval

    Cite this

    Iakovidou, C., Anagnostopoulos, N., Lux, M., Christodoulou, K., Boutalis, Y., & Chatzichristofis, S. A. (2019). Composite Description Based on Salient Contours and Color Information for CBIR Tasks. IEEE Transactions on Image Processing, 28(6), 3115-3129. [8626513]. https://doi.org/10.1109/TIP.2019.2894281
    Iakovidou, C. ; Anagnostopoulos, N. ; Lux, M. ; Christodoulou, K. ; Boutalis, Y. ; Chatzichristofis, S. A. / Composite Description Based on Salient Contours and Color Information for CBIR Tasks. In: IEEE Transactions on Image Processing. 2019 ; Vol. 28, No. 6. pp. 3115-3129.
    @article{90e269cec7694e1c97efa6fff8acdb47,
    title = "Composite Description Based on Salient Contours and Color Information for CBIR Tasks",
    abstract = "This paper introduces a novel image descriptor for content-based image retrieval tasks that integrates contour and color information into a compact vector. Loosely inspired by the human visual system and its mechanisms in efficiently identifying visual saliency, operations are performed on a fixed lattice of discrete positions by a set of edge detecting kernels that calculate region derivatives at different scales and orientation. The description method utilizes a weighted edge histogram where bins are populated on the premise of whether the regions contain edges belonging to the salient contours, while the discriminative power is further enhanced by integrating regional quantized color information. The proposed technique is both efficient and adaptive to the specifics of each depiction, while it does not need any training data to adjust parameters. An experimental evaluation conducted on seven benchmarking datasets against 13 well known global descriptors along with SIFT, SURF implementations (both in VLAD and BOVW), highlight the effectiveness and efficiency of the proposed descriptor.",
    keywords = "global features, Image retrieval, multimedia retrieval",
    author = "C. Iakovidou and N. Anagnostopoulos and M. Lux and K. Christodoulou and Y. Boutalis and Chatzichristofis, {S. A.}",
    year = "2019",
    month = "6",
    day = "1",
    doi = "10.1109/TIP.2019.2894281",
    language = "English",
    volume = "28",
    pages = "3115--3129",
    journal = "IEEE Transactions on Image Processing",
    issn = "1057-7149",
    number = "6",

    }

    Iakovidou, C, Anagnostopoulos, N, Lux, M, Christodoulou, K, Boutalis, Y & Chatzichristofis, SA 2019, 'Composite Description Based on Salient Contours and Color Information for CBIR Tasks', IEEE Transactions on Image Processing, vol. 28, no. 6, 8626513, pp. 3115-3129. https://doi.org/10.1109/TIP.2019.2894281

    Composite Description Based on Salient Contours and Color Information for CBIR Tasks. / Iakovidou, C.; Anagnostopoulos, N.; Lux, M.; Christodoulou, K.; Boutalis, Y.; Chatzichristofis, S. A.

    In: IEEE Transactions on Image Processing, Vol. 28, No. 6, 8626513, 01.06.2019, p. 3115-3129.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Composite Description Based on Salient Contours and Color Information for CBIR Tasks

    AU - Iakovidou, C.

    AU - Anagnostopoulos, N.

    AU - Lux, M.

    AU - Christodoulou, K.

    AU - Boutalis, Y.

    AU - Chatzichristofis, S. A.

    PY - 2019/6/1

    Y1 - 2019/6/1

    N2 - This paper introduces a novel image descriptor for content-based image retrieval tasks that integrates contour and color information into a compact vector. Loosely inspired by the human visual system and its mechanisms in efficiently identifying visual saliency, operations are performed on a fixed lattice of discrete positions by a set of edge detecting kernels that calculate region derivatives at different scales and orientation. The description method utilizes a weighted edge histogram where bins are populated on the premise of whether the regions contain edges belonging to the salient contours, while the discriminative power is further enhanced by integrating regional quantized color information. The proposed technique is both efficient and adaptive to the specifics of each depiction, while it does not need any training data to adjust parameters. An experimental evaluation conducted on seven benchmarking datasets against 13 well known global descriptors along with SIFT, SURF implementations (both in VLAD and BOVW), highlight the effectiveness and efficiency of the proposed descriptor.

    AB - This paper introduces a novel image descriptor for content-based image retrieval tasks that integrates contour and color information into a compact vector. Loosely inspired by the human visual system and its mechanisms in efficiently identifying visual saliency, operations are performed on a fixed lattice of discrete positions by a set of edge detecting kernels that calculate region derivatives at different scales and orientation. The description method utilizes a weighted edge histogram where bins are populated on the premise of whether the regions contain edges belonging to the salient contours, while the discriminative power is further enhanced by integrating regional quantized color information. The proposed technique is both efficient and adaptive to the specifics of each depiction, while it does not need any training data to adjust parameters. An experimental evaluation conducted on seven benchmarking datasets against 13 well known global descriptors along with SIFT, SURF implementations (both in VLAD and BOVW), highlight the effectiveness and efficiency of the proposed descriptor.

    KW - global features

    KW - Image retrieval

    KW - multimedia retrieval

    UR - http://www.scopus.com/inward/record.url?scp=85065436756&partnerID=8YFLogxK

    U2 - 10.1109/TIP.2019.2894281

    DO - 10.1109/TIP.2019.2894281

    M3 - Article

    VL - 28

    SP - 3115

    EP - 3129

    JO - IEEE Transactions on Image Processing

    JF - IEEE Transactions on Image Processing

    SN - 1057-7149

    IS - 6

    M1 - 8626513

    ER -