On Optimized Color Image Coding Using Correlation of Primary Colors

  • Eyal Braunstain
  • Moshe Porat
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8156)


The RGB color primaries in natural images are characterized by a high degree of inter-correlation. Many compression algorithms use this information redundancy to reduce the amount of bits required for coding, by transforming the color information to a decorrelated color space - such as YUV. The human visual system is more sensitive to luminance than chrominance components, so more bits are allocated to luminance. We examine a different approach, by expressing two of the color components (termed subordinate colors) as a functional of the other color component (termed base color). Unlike some compression algorithms (e.g. JPEG) that perform the analysis on NxN blocks in the image, we utilize segmentation by Region Growing in both gray level and color (RGB) images. We suggest a method for selection of optimal base color for each region separately. The proposed approach could be useful for color compression and progressive transmission applications.


Color compression Correlation Polynomial approximation 


  1. 1.
    Goffman-Vinopal, L., Porat, M.: Color image compression using inter-color correlation. In: Proceedings of the International Conference on Image Processing 2002, vol. 2, pp. II-353–II-356 (2002)Google Scholar
  2. 2.
    Kotera, H., Kanamori, K.: A novel coding algorithm for representing full color images by a single color image. J. Imag. Technol. 16, 142–152 (1990)Google Scholar
  3. 3.
    Brice, C.R., Fennema, L.: Scene analysis using regions. Artif. Intell. (1), 205–226 (1970)Google Scholar
  4. 4.
    Kunt, M., Ikonomopoulos, A., Kocher, M.: Second-generation image-coding techniques. Proc. IEEE 73(4), 549–573 (1985)CrossRefGoogle Scholar
  5. 5.
    Vedaldi, A., Soatto, S.: Quick Shift and Kernel Methods for Mode Seeking. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part IV. LNCS, vol. 5305, pp. 705–718. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  6. 6.
    Wang, D.C.C., Vaganucci, A.H.: Gradient inverse weighted smoothing scheme and evaluation of performance. Comput. Graph. Image Process. 15 (1981)Google Scholar
  7. 7.
    Roterman, Y., Porat, M.: Color image coding using regional correlation of primary colors. Image and Vision Computing 25, 637–651 (2007)CrossRefGoogle Scholar
  8. 8.
    Freeman, H.: On the encoding of arbitrary geometric configuration. IRE Trans. Electron. Comput. EC-10, 260–268 (1961)Google Scholar
  9. 9.
    Porat, M., Zeevi, Y.Y.: The generalized Gabor scheme of image representation in biological and machine vision. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI) 10(44), 452–468 (1988)CrossRefzbMATHGoogle Scholar
  10. 10.
    Eldar, Y., Lindenbaum M., Porat M., Zeevi Y.Y.: The farthest point strategy for progressive image sampling. IEEE Transactions on Image Processing 6(9), 1305–1315 (1997)Google Scholar
  11. 11.
    Wallace, G.K.: The JPEG still picture compression standard. IEEE Transactions on Consumer Electronics 38(1), xviii-xxxiv (1992)Google Scholar
  12. 12.
    Christopoulos, C., Skodras, A., Ebrahimi, T.: The JPEG 2000 Still Image Coding System: An Overview. IEEE Transactions on Consumer Electronics 46(4), 1103–1127 (2000)CrossRefGoogle Scholar
  13. 13.
    Nemirovsky, S., Porat, M.: On texture and image interpolation using Markov models. Signal Processing: Image Communication 24(3), 139–157 (2009)CrossRefGoogle Scholar
  14. 14.
    Genossar, T., Porat, M.: Optimal bi-orthonormal approximation of signals. IEEE Transactions on Systems, Man and Cybernetics 22(3), 449–460 (1992)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Eyal Braunstain
    • 1
  • Moshe Porat
    • 2
  1. 1.Department of Biomedical EngineeringTechnion - Israel Institute of TechnologyHaifaIsrael
  2. 2.Department of Electrical EngineeringTechnion - Israel Institute of TechnologyHaifaIsrael

Personalised recommendations