Content-based image collection profiling and comparison via self-organised maps

ABSTRACT

The research of content-based image retrieval techniques has been focused on extracting effective low level visual features for indexing and enabling query of individual images by efficient feature matching. In this paper, the content-based approach is extended towards the problem of multimedia collection profiling and comparison. We propose to carry out visual feature clustering using the Kohonen self-organized map algorithm, and then apply distance measures on the generated feature maps to evaluate their similarity. Apart from the conventional Hausdorff distance, other distance measures have been implemented and found to perform better in our case study.

References

1. {1} C. Carson, M. Thomas, S. Belongie, and et al. Blobworld: A system for region-based image indexing and retrieval. In Proc. Int. Conf. Visual Inf. Sys., pages 509-516, 1999.]]
2. {2} D. Deng and N. Kasabov. On-line pattern analysis by evolving self-organizing maps. Neurocomputing, 51:87-103, 2003.]]
3. {3} T. Eiter and H. Mannila. Distance measures for point sets and their computation. Acta Informica, 34:109-133, 1997.]]
4. {4} B. Fritzke. A growing neural gas network teams topologies. In Proc. of NIPS, 1994.]]
5. {5} S. Haykin. Neural Networks: A Comprehensive Foundation. Prentice Hall, second edition, 1999.]]
6. {6} K. Hirata and S. Mukherjea. Integration of image matching and classification for multimedia navigation. Multimedia Tools and Application, 11:295-309, 2000.]]
7. {7} T. Honkela. Comparisons of self-organized word category maps. In Proceedings of WSOM97, Workshop on Self-Organizing Maps, pages 298-303. Helsinki University of Technology, Neulal Networks Research Centre, Espoo, Finland, 1997.]]
8. {8} J. Huang, S. Kumar, M. Mitra, and W. Zhu. Spatial color indexing and applications. In Proc. of ICCV, pages 602-607, 1998.]]
9. {9} S. Kaski and K. Lagus. Comparing self-organizing maps. In J. Vorbruggen and B. Sendhoff, editors, Proceedings of ICANN96 International Conference on Artificial Neural Networks, volume 1112 of Lecture Notes in Computer Science, pages 809-814. Springer, Berlin, 1996.]]
10. {10} T. Kohonen. Self-organizing Maps. Springer-Verlag, second edition, 1997.]]
11. {11} J. Laaksonen, M. Koskela, and E. Oja. Content-based image retrieval using self-organizing maps. In Visual Information and Information Systems, pages 541-548, 1999.]]
12. {12} A. Mukerjee, K. Gupta, S. Nautiyal, and et al. Conceptual description of visual scenes from linguistic models. Image and Vision Computing, 18:173-187, 2000.]]
13. {13} G. Pass and R. Zabih. Comparing images using joint histograms. Multimedia Systems, 7(3):234-240, 1999.]]
14. {14} J. Puzicha, Y. Rubner, C. Tomasi, and J. Buhmann. Empirical evaluation of dissimilarity measures for color and texture. In Proceedings the IEEE International Conference on Computer Vision(ICCV-1999), pages 1165-1173, 1999.]]
15. {15} A. Rauber and D. Merkl. The somlib digital library system. In Proc. of European Conference on Digital Libraries, pages 323-342, 1999.]]
16. {16} H. Ritter. Asymptotic level density for a class of vector quantization processes. IEEE Trans. Neural Networks, 2:173-175, 1991.]]
17. {17} Y. Rubner, C. Tomasi, and L. Guibas. A metric for distributions with applications to image databases. In Proc. of IEEE ICCV, pages 59-66, 1998.]]
18. {18} S. Santini and R. Jain. Beyond query by example. In Proc. of ACM Multimedia 98, pages 345-350, 1998.]]
19. {19} A. Smeulders, M. Worring, S. Santini, A. Gupta, and J. R. Content-based image retrieval at the end of the early years. IEEE Transaction on Pattern Analysis and Machine Intelligence, 22(12):1349-1380, 2000.]]
20. {20} J. Smith and S. Chang. Visualseek: a fully automated content-base image query system. In Proc. of ACM Multimedia 96, pages 87-98, 1996.]]

Index Terms

• Computing methodologies

  • Artificial intelligence

    • Computer vision

  • Machine learning

    • Learning paradigms

      • Supervised learning

        • Supervised learning by classification

    • Machine learning approaches

      • Classification and regression trees

      • Neural networks