Image retrieval by multi-scale illumination invariant Indexing

  • Theo Gevers
  • Arnold W.M. Smeulders
Invited Talk
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1464)


The purpose is to arrive at image retrieval invariant to a substantial change in illumination.

We will extend the theory that we have recently proposed on illumination invariant color models [6]. Then, a multi-scale image representation is produced by applying Gaussian derivatives at different scale levels on the illumination invariant color models. In this way, a multi-dimensional multi-scale image index is obtained which is illumination-independent and invariant under the group of rotations in the image domain. The multi-scale image representation is taken as input for image retrieval by query by example (i.e. an example image is given by the user) and image retrieval by arranging the image database as a binary tree (i.e. no example image is given is available).

Experiments have been conducted on a database consisting of 500 images taken from multicolored man-made objects in real world scenes. From the experimental results it can be observed that image retrieval by multi-scale invariant indexing provides high retrieval accuracy even under spatially and spectrally varying illumination.


image retrieval dichromatic reflection reflectance properties photometric color invariants Gaussian derivatives scale-space multi-scale invariant indexing query by example decision trees 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Finlayson, G. D., Drew M. S., and Funt B. V.: Spectral Sharpening: Sensor Transformations for improved Color Constancy. J. Opt. Soc. Am. 11(5) (1994) 1553–1563Google Scholar
  2. 2.
    Finlayson, G. D., Chatterjee S. S., and Funt B. V.: Color Angular Indexing. ECCV96 II (1996) 16–27Google Scholar
  3. 3.
    Forsyth, D.: A Novel Algorithm for Color Constancy. International Journal of Computer Vision Vol. 5 1990 5–36Google Scholar
  4. 4.
    Funt, B. V. and Drew, M. S.: Color Constancy Computation in Near-Mondrian Scenes. In Proceedings of the CVPR IEEE Computer Society Press 1988 544–549Google Scholar
  5. 5.
    Funt, B. V. and Finlayson, G. D.: Color Constant Color Indexing. IEEE PAMI 17(5) 1995 522–529Google Scholar
  6. 6.
    Gevers, T. and Smeulders, A. W. M.: Image Indexing using Composite Color and Shape Invariant Features. ICCV Bombay India (1998)Google Scholar
  7. 7.
    Hartigan, J. A.: Clustering Algorithms. John Wiley and Sons U.S.A (1975)Google Scholar
  8. 8.
    Healey, G. and Slater D.: Global Color Constancy: Recognition of Objects by Use of Illumination Invariant Properties of Color Distributions. J. Opt. Soc. Am. A Vol. 11 No. 11 (1995) 3003–3010Google Scholar
  9. 9.
    Koenderink, J. J. and van Doorn A. J.: Representation of Local Geometry in the Visual System. Biological Cybernetics No. 55 (1987) 367–375Google Scholar
  10. 10.
    Land, E. H. and McCann, J. J.: Lightness and Retinex Theory. J. Opt. Soc. Am. Vol. 61 (1971) 1–11Google Scholar
  11. 11.
    Lee H.-C., Breneman E. J. and Schulte C. P.: Modeling Light Reflection for Computer Color Vision. IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 12 No. 3 (1990) 402–409Google Scholar
  12. 12.
    Levkowitz, H. and Herman G. T.: GLHS: A Generalized Lightness, Hue, and Saturation Color Model. CVGIP: Graphical Models and Image Processing Vol. 55 No. 4 (1993) 271–285Google Scholar
  13. 13.
    Nayar, S. K. and Bolle, R. M.: Reflectance Based Object Recognition. International Journal of Computer Vision Vol. 17 No. 3 1996 219–240Google Scholar
  14. 14.
    Shafer, S. A.: Using Color to Separate Reflection Components. COLOR Res. Appl. 10(4) (1985) 210–218Google Scholar
  15. 15.
    D. Slater and G. Healey: The Illumination-invariant Recognition of 3D Objects Using Local Color Invariants. IEEE Trans. PAMI 18(2) (1996)Google Scholar
  16. 16.
    Swain, M. J. and Ballard, D. H.: Color Indexing. International Journal of Computer Vision Vol. 7 No. 1 1991 11–32Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Theo Gevers
    • 1
  • Arnold W.M. Smeulders
    • 1
  1. 1.Intelligent Sensory Information SystemsUniversity of AmsterdamSJ AmsterdamThe Netherlands

Personalised recommendations