Image Segmentation & the use of Genetic Algorithms for Optimising Parameters and Algorithm choice

  • P. G. Ducksbury
Conference paper


One of the difficulties that has been apparent in applying image processing and understanding algorithms is that of the optimal choice of parameters and the algorithms themselves. Firstly we must select an algorithm and secondly the actual parameters that are required by that algorithm. It is also the case that using a chosen algorithm on a different image class yields results of a totally different quality, we have considered three image classes, namely infra-red linescan, Russian satellite and SPOT imagery. We have explored the use of genetic algorithms for the purpose of parameter and algorithm selection and will show how the approach can successfully obtain results which in the past have tended to be obtained somewhat heuristically. Once a reliable region has been obtained then we can represent its shape using a curvature scale space description.The main application of this work will be in the area of image databases.


Genetic Algorithm Image Class Roulette Wheel Selection Infeasible Point Curvature Scale Space 
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]
    P. G. Ducksbury, Parallel Texture Region Segmentation using a Pearl Bayes Network, British Machine Vision Conference, BMVC-93, University of Surrey, Sep 1993, pp 187–196.Google Scholar
  2. [2]
    P. G. Ducksbury, M. J. Varga, Region Based Image Content Descriptors and Representation, 6 th IEE Int. Conf. on Image Processing & its Applications, Trinity College, Dublin, July 1997.Google Scholar
  3. [3]
    Zhi-Yan Xie, Multi-scale Analysis and Texture Segmentation, PhD Thesis, Dept of Eng Science, University of Oxford, 1994.Google Scholar
  4. [4]
    P. L. Palmer, H. Dabis, J. Kittler, A performance measure for boundary detection algorithms, Computer Vision and Image Understanding, Vol 63, No 3, pp 476–494, 1996.CrossRefGoogle Scholar
  5. [5]
    D. E. Goldberg, Genetic Algorithms in Search, Optimisation and Machine Learning, Addison Wesley, 1989.MATHGoogle Scholar
  6. [6]
    F. Mokhtarian, Silhouette-Based isolated object recognition through curvature scale space, IEEE Trans PAMI, vol 17, no 5, May 1995.Google Scholar
  7. [7]
    P.G. Ducksbury, M.J. Varga, P.K. Kent, S. Foulkes, D.M. Booth, Genetic algorithms for automatic algorithm and parameter selection in ATR applications’, SPIE Aerosense-98, Conf 3371, Orlando, 13–17th April, 1998.Google Scholar

Copyright information

© Springer-Verlag London Limited 1998

Authors and Affiliations

  • P. G. Ducksbury
    • 1
  1. 1.Defence Evaluation & Research AgencyMalvernUK

Personalised recommendations