Evolving Approximate Image Filters

  • Simon Colton
  • Pedro Torres
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5484)


Image filtering involves taking a digital image and producing a new image from it. In software packages such as Adobe’s Photoshop, image filters are used to produce artistic versions of original images. Such software usually includes hundreds of different image filtering algorithms, each with many fine-tuneable parameters. While this freedom of exploration may be liberating to artists and designers, it can be daunting for less experienced users. Photoshop provides image filter browsing technology, but does not yet enable the construction of a filter which produces a reasonable approximation of a given filtered image from a given original image. We investigate here whether it is possible to automatically evolve an image filter to approximate a target filter, given only an original image and a filtered version of the original. We describe a tree based representation for filters, the fitness functions and search techniques we employed, and we present the results of experimentation with various search setups. We demonstrate the feasibility of evolving image filters and suggest new ways to improve the process.


Original Image Random Generation Colour Histogram Image Filter Colourful Target 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Behrenbruch, C., Petroudi, S., Bond, S., Declerck, J., Leong, F., Brady, J.: Image filtering techniques for medical image post-processing: an overview. British Journal of Radiology 77(2)Google Scholar
  2. 2.
    Harding, S.: Evolution of image filters on graphics processor units using Cartesian genetic programming. In: IEEE Congress on Evolutionary Computation (2008)Google Scholar
  3. 3.
    Machado, P., Cardoso, A.: All the truth about NEvAr. App. Int. 16, 101–118 (2002)CrossRefzbMATHGoogle Scholar
  4. 4.
    Machado, P., Dias, A., Duarte, N., Cardoso, A.: Giving Colour to Images. In: Proceedings of the AISB 2002, Symposium on Artificial Intelligence and Creativity in Arts and Science (2002)Google Scholar
  5. 5.
    Poli, R., Langdon, W.: Schema Theory for Genetic Programming with One-Point Crossover and Point Mutation. Evolutionary Computation 6(3), 231–252 (1998)CrossRefGoogle Scholar
  6. 6.
    Sekanina, L., Martínek, T.: Evolving image operators directly in hardware. In: Proc. of Genetic and Evol. Computation for Image Processing and Analysis (2007)Google Scholar
  7. 7.
    Sims, K.: Artificial evolution for computer graphics. In: Proceedings of SIGGRAPH 1991 (1991)Google Scholar
  8. 8.
    Smith, S., Leggett, S., Tyrrell, A.: An implicit context representation for evolving image processing filters. In: Rothlauf, F., et al. (eds.) EvoWorkshops 2005. LNCS, vol. 3449, pp. 407–416. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  9. 9.
    Torres, P., Colton, S., Rüger, S.: Experiments in example-based image filter retrieval. In: Proceedings of the Cross-Media Workshop (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Simon Colton
    • 1
  • Pedro Torres
    • 1
  1. 1.Computational Creativity Group Department of ComputingImperial College LondonUK

Personalised recommendations