Invisible modification of the palette color image enhancing lossless compression

  • Jaroslav Fojtík
  • Václav Hlaváč
Poster Papers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1451)


We address the problem of pseudocolor image compression. Image values represent indices into a look up table (palette). Due to quantization, the neighbouring pixel values (indices) change too much. This deteriorates performance of both lossless and lossy image compression methods. We suggest a preprocessing phase that (a) analyses statistics of the adjacency relations of index values, (b) performs palette optimization, and (c) permutes indices to palette to achieve more smooth image. The smoother image causes that the lossless image compression methods yield less output data. The task to optimally permute palette indices is a NP complete combinatorial optimization. Instead of checking all possibilities, we suggest a reasonable: initial guess and a fast suboptimal hill climbing optimization. The proposed permutation of indices should enhance performance of most lossless compression method used after it. To our knowledge, the proposed reordering followed by our own nonlinear compression technique [IIF97b, HF97a] yields the best compression. Experiments with various images show that the indices reordering provides data savings from 10% to 50%.


Image Compression Gray Level Image Lossless Compression Adjacency Relation Lossless Image Compression 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. [AL93]
    Zaccarin André and Bede Liu. A novel approach for coding color quantized images. IEEE Transactions on Image Processing, 2(4):442–453, October 1993.CrossRefGoogle Scholar
  2. [FI198]
    Jaroslav Fojtík and Václav Hlaváč Invisible modification of palette color image for increasing compression ratio of lossless compression methods. Technical l1eport K335/98/159, Czech Technical University, Faculty of Electrical Engineering, Karlovo Niměsti 13, Prague 2, May 1998Google Scholar
  3. [Ry93]
    Michael Frydrych. Image compression. Master's thesis, Charles University, Faculty of Mathematics Physics, Prague, Czech Reepublic, 1993.Google Scholar
  4. [HF97a]
    V. Hlaváč and J. Fojtík Adaptive non-linear predictor for lossless image compression. In G. Sommer, K. Daniilidis, and J. Pauli, editors, Proceedings of the conference Computer Analysis of Images and Patterns'97, Kiel, Germany, pages 279–288. Springer-Verlag, LNCS 1296, September 1997.Google Scholar
  5. [HF97b]
    V. Hlaváč and J. Fojtík. Predictor based on frequency analysis of the local configurations used for lossless image compression. In Proceedings of the 1st IAPR TC1 workshop on Statistical Techniques in Pattern Recognition, Prague, Czech Republic, June 9–11, 1997, pages 73–78, Prague, Czech Republic, June 1997. Institute of Information Theory and Automation, Czech Acadeiny of Sciences.Google Scholar
  6. [HS94]
    Andrew C. Hadenfeldt and Khaid Sayood. Compression of color-mapped images. IEEE Transactions on Geoscience and Remote Sensing, 32(3):534–541, May 1994.CrossRefGoogle Scholar
  7. [HS92]
    R. M. Haralick and L. G. Shapiro. Computer and Robot Vision, Volume I. Addison Wesley, Reading, Ma., 1992.Google Scholar
  8. [MV96]
    Nasir D. Memon and Ayalur Venkateswaran. On ordering color maps for lossless predictive coding. IEEE Transactions on Image Processing, 5(11):1522–1527, November 1996.CrossRefGoogle Scholar
  9. [Sch89]
    M.I. Schlesinger. Matematiceskie sredstva obrabotki izobrazenij, in Russian, (Mathematic tools for image processing). Naukova Dumka, Kiev, Ukraine, 1989.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Jaroslav Fojtík
    • 1
  • Václav Hlaváč
    • 1
  1. 1.Faculty of Electrical Engineering Center for Machine PerceptionCzech Technical UniversityKarlovo náměstí 13Czech Republic

Personalised recommendations