Scale Space Hierarchy of Segments

  • Haruhiko Nishiguchi
  • Atsushi Imiya
  • Tomoya Sakai
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5702)


In this paper, we develop a segmentation algorithm using configurations of singular points in the linear scale space. We define segment edges as a zero-crossing set in the linear scale space using the singular points. An image in the linear scale space is the convolution of the image and the Gaussian kernel. The Gaussian kernel of an appropriate variance is a typical presmoothing operator for segmentation. The variance is heuristically selected using statistics of images such as the noise distribution in images. The variance of the kernel is determined using the singular point configuration in the linear scale space, since singular points in the linear scale space allow the extraction of the dominant parts of an image. This scale selection strategy derives the hierarchical structure of the segments.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Zhao, N.-Y., Iijima, T.: Theory on the method of determination of view-point and field of vision during observation and measurement of figure. IECE Japan, Trans. D  J68-D, 508–514 (1985) (in Japanese)Google Scholar
  2. 2.
    Zhao, N.-Y., Iijima, T.: A theory of feature extraction by the tree of stable view-points. IECE Japan, Trans. D J68-D, 1125–1135 (1985) (in Japanese) Google Scholar
  3. 3.
    Iijima, T.: Pattern Recognition, Corona-sha, Tokyo (1974) (in Japanese)Google Scholar
  4. 4.
    Witkin, A.P.: Scale space filtering. In: Proc. of 8th IJCAI, pp. 1019–1022 (1993)Google Scholar
  5. 5.
    Lindeberg, T.: Scale-Space Theory in Computer Vision. Kluwer, Boston (1994)Google Scholar
  6. 6.
    Lindeberg, T.: Feature detection with automatic selection. IJCV 30, 79–116 (1998)CrossRefGoogle Scholar
  7. 7.
    ter Haar Romeny, B.M.: Front-End Vision and Multi-Scale Image Analysis Multi-scale Computer Vision Theory and Applications, written in Mathematica. Springer, Berlin (2003)Google Scholar
  8. 8.
    Weicker, J.: Anisotropic Diffusion in Image Processing. Teubner, Stuttgart (1998)Google Scholar
  9. 9.
    Imiya, A., Sugiura, T., Sakai, T., Kato, Y.: Temporal structure tree in digital linear scale space. In: Griffin, L.D., Lillholm, M. (eds.) Scale-Space 2003. LNCS, vol. 2695, pp. 356–371. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  10. 10.
    Pelillo, M., Siddiqi, K., Zucker, S.W.: Matching hierarchical structures using association graphs. PAMI 21, 1105–1120 (1999)Google Scholar
  11. 11.
    Yuille, A.L., Poggio, T.: Scale space theory for zero crossings. PAMI 8, 15–25 (1986)MATHGoogle Scholar
  12. 12.
    Kuijper, A., Florack, L.M.J., Viergever, M.A.: Scale space hierarchy. Journal of Mathematical Imaging and Vision 18, 169–189 (2003)MATHCrossRefMathSciNetGoogle Scholar
  13. 13.
    Kuijper, A., Florack, L.M.J.: The hierarchical structure of images. IEEE Trans. Image Processing 12, 1067–1079 (2003)CrossRefMathSciNetGoogle Scholar
  14. 14.
    Enomoto, H., Yonezaki, N., Watanabe, Y.: Application of structure lines to surface construction and 3-dimensional analysis. In: Fu, K.-S., Kunii, T.L. (eds.) Picture Engineering, pp. 106–137. Springer, Berlin (1982)Google Scholar
  15. 15.
    Krueger, W.M., Phillips, K.: The geometry of differential operator with application to image processing. PAMI 11, 1252–1264 (1989)Google Scholar
  16. 16.
    Canny, J.: A computational approach to edge detection. PAMI 8, 679–698 (1986)Google Scholar
  17. 17.
    Najman, L., Schmitt, M.: Watershed of a continuous function. Signal Processing 38, 99–112 (1994)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Haruhiko Nishiguchi
    • 1
  • Atsushi Imiya
    • 2
  • Tomoya Sakai
    • 2
  1. 1.School of Science and TechnologyChiba UniversityJapan
  2. 2.Institute of Media and Information TechnologyChiba University, JapanChibaJapan

Personalised recommendations