Human-Centered Content-Based Image Retrieval

Egon van den Broek

    Research output: ThesisPhD Thesis - Research UT, graduation UT

    461 Downloads (Pure)

    Abstract

    Retrieval of images that lack a (suitable) annotations cannot be achieved through (traditional) Information Retrieval (IR) techniques. Access through such collections can be achieved through the application of computer vision techniques on the IR problem, which is baptized Content-Based Image Retrieval (CBIR). In contrast with most purely technological approaches, the thesis Human-Centered Content-Based Image Retrieval approaches the problem from a human/user centered perspective. Psychophysical experiments were conducted in which people were asked to categorize colors. The data gathered from these experiments was fed to a Fast Exact Euclidean Distance (FEED) transform (Schouten & Van den Broek, 2004), which enabled the segmentation of color space based on human perception (Van den Broek et al., 2008). This unique color space segementation was exploited for texture analysis and image segmentation, and subsequently for full-featured CBIR. In addition, a unique CBIR-benchmark was developed (Van den Broek et al., 2004, 2005). This benchmark was used to explore what and how several parameters (e.g., color and distance measures) of the CBIR process influence retrieval results. In contrast with other research, users judgements were assigned as metric. The online IR and CBIR system Multimedia for Art Retrieval (M4ART) (URL: http://www.m4art.org) has been (partly) founded on the techniques discussed in this thesis. References: - Broek, E.L. van den, Kisters, P.M.F., and Vuurpijl, L.G. (2004). The utilization of human color categorization for content-based image retrieval. Proceedings of SPIE (Human Vision and Electronic Imaging), 5292, 351-362. [see also Chapter 7] - Broek, E.L. van den, Kisters, P.M.F., and Vuurpijl, L.G. (2005). Content-Based Image Retrieval Benchmarking: Utilizing Color Categories and Color Distributions. Journal of Imaging Science and Technology, 49(3), 293-301. [see also Chapter 8] - Broek, E.L. van den, Schouten, Th.E., and Kisters, P.M.F. (2008). Modeling Human Color Categorization. Pattern Recognition Letters, 29(8), 1136-1144. [see also Chapter 5] - Schouten, Th.E. and Broek, E.L. van den (2004). Fast Exact Euclidean Distance (FEED) transformation. In J. Kittler, M. Petrou, and M. Nixon (Eds.), Proceedings of the 17th IEEE International Conference on Pattern Recognition (ICPR 2004), Vol 3, p. 594-597. August 23-26, Cambridge - United Kingdom. [see also Appendix C]
    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • Radboud University Nijmegen
    Supervisors/Advisors
    • Vuurpijl, L.G., Advisor, External person
    • Schouten, Th.E., Advisor, External person
    • de Weert, Ch.M.M., Supervisor, External person
    Thesis sponsors
    Award date21 Sept 2005
    Place of PublicationNijmegen
    Publisher
    Print ISBNs90-901-9730-3
    Publication statusPublished - 21 Sept 2005

    Keywords

    • cognitive computer vision
    • IR-78703
    • Benchmark
    • HMI-CI: Computational Intelligence
    • Content-Based Image Retrieval (CBIR)
    • EWI-20850
    • Human-Centered
    • Machine Learning
    • Texture
    • shape
    • Segmentation
    • Information Retrieval (IR)
    • Image Processing
    • HMI-HF: Human Factors
    • Pattern Recognition
    • objects
    • HMI-MR: MULTIMEDIA RETRIEVAL
    • Color

    Fingerprint

    Dive into the research topics of 'Human-Centered Content-Based Image Retrieval'. Together they form a unique fingerprint.

    Cite this