Analysis and Comparison of different Genetic Models for the Clustering problem in Image Analysis

  • Rita Cucchiara


This paper presents several genetic approaches to the clustering problem of N elements in an n-dimensional Feature Space. This process has been applied in an Image Analysis context in order to divide a set of objects into a fixed number of groups, dependending on their characteristics. The partitioning models are based on very general issues so they can be used in many different clustering applications, as well as real objects grouping. The genetic paradigm has been choisen because the cluster Solution Space has to be explored without any ‘a priori’ or heuristic knowledge and also because the performed parallel search can elude the relevant number of local minima in the solution optimisation. Different cluster models and genetic operators have been analysed in order to exploit the genetic algorithm power in an Image Analysis environment. A performance comparison between solutions is shown, using several chromosome codes and genetic operators.


Genetic Algorithm Crossover Operator Cluster Problem Uniform Crossover Genetic Paradigm 
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]
    J.M. Jolion, P. Meer, S. Batouche ‘Robust Clustering with Applications in Computer Vision’, IEEE Trans. on P.A.M.I., vol. 13, no. 8, Aug. 1991Google Scholar
  2. [2]
    D. Vernon ‘Machine Vision: Automated Visual Inspection and Robot Vision’ Prentice Hall 1991Google Scholar
  3. [3]
    D.E. Goldberg, ‘Genetic Algorithms, in search Optimization & Machine Learning’, Addison-Wesely 1989Google Scholar
  4. [4]C.
    Alippi, R. Cucchiara, ‘Cluster Partitioning in Image Analysis Classification: a Genetic Algorithm Approach,’ Proc. of the IEEE Int. Conf. COMPEURO 92, 1992 The Hague.Google Scholar
  5. [5]
    K.A._De Jong ‘Adaptive System Design: a Genetic Approach’, IEEE Trans. on Systems, Man & Cybernetics SMC-10, 9, pp.566-574Google Scholar
  6. [6]
    M.D. Vose ‘Generalizing the notion of schema in genertic algorithms’ Artificial Intelligence 1991 n. 50 pp 385-396Google Scholar
  7. [7]
    D.E. Goldberg, ‘Sizing Population for Serial and Parallel Genetic Algorithms’, Proc. of the Third Int. Conf. on G.A. and their Applications, pp. 70-79, Morgan Kaufmann 1989Google Scholar
  8. [8]
    J.J. Grefenstette, ‘Optimization of Control Parameters for Genetic Algorithms’, IEEE Trans. on Systems, Man & Cybernetics SMC-16, 1, 1986Google Scholar
  9. [9]G.
    Syswerda ‘Uniform Crossover in Genetic Algorithms’, Proc. on the 3 Int. Conf. on G.A. and their Appl. pp. 2-9, 1989Google Scholar
  10. [10]
    T.C. Fogarty ‘Varying the Probability of Mutation in the Genetic Algorithm’, Proc. of the Third Int. Conf. on G.A. and their Applications, pp. 104-109, 1989Google Scholar

Copyright information

© Springer-Verlag/Wien 1993

Authors and Affiliations

  • Rita Cucchiara
    • 1
  1. 1.Istituto di IngegneriaUniversita’ degli Studi di FerraraFerraraItaly

Personalised recommendations