Advertisement

Comparison of Two Evolutionary Approaches for the Polygonal Approximation of Digital Curves

  • Paola Berenice Alvarado-Velazco
  • Victor Ayala-Ramirez
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 175)

Abstract

In this paper, we address the approximation of digital curves using straight-line segments. Our objective is to compare the performance of two population-based evolutionary algorithms: an evolutionary programming approach and a variable length chromosome genetic algorithm to solve the polygonal approximation problem.We describe the main elements of the methods under comparison and we show the results of the tests executed on a dataset comprising curves that exhibit a range of conditions with respect to two main features: openness and straightness. Our experiments show that the evolutionary programming based technique is faster and more accurate than the genetic algorithm based approach.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kolesnikov, A.: Fast algorithm for error-bounded compression of digital curves. In: Proc. of the 2010 IEEE Int. Conf. on Image Processing, Hong Kong, pp. 1453–1456 (2010)Google Scholar
  2. 2.
    Wang, Z., Li, H.Y.: User-controlled geometric feature preserving simplification. In: Fifth Interntional Conference on Frontier of Computer Science and Technology, pp. 335–340 (2010)Google Scholar
  3. 3.
    Sirakov, N., Swift, J., Mlsna, P.: Image database query using shape-based boundary descriptors. In: Proc. of International Conference on Intelligent Systems Design and Applications (ISDA 2003), pp. 373–382 (2003)Google Scholar
  4. 4.
    Perez, J., Vidal, E.: Optimum polygonal approximation of digitized curves. Pattern Recognition Letters 15, 743–750 (1994)MATHCrossRefGoogle Scholar
  5. 5.
    Tsai, Y.H.: Fast polygonal approximation based on genetic algorithms. In: Proceedings of the 5th IEEE/ACIS International Conference on Computer and Information Science and 1st IEEE/ACIS International Workshop on Component-Based Software Engineering, Software Architecture and Reuse, pp. 322–326 (2006)Google Scholar
  6. 6.
    Traver, V.J., Recatalá, G., Iñesta, J.M.: Exploring the performance of genetic algorithms as polygonal approximators. In: 15th International Conference on Pattern Recognition, pp. 766–769 (2000)Google Scholar
  7. 7.
    Guanghui, L., Chuanbo, C.: A new approach for polygonal approximation of shape contours using genetic algorithms. In: Second IEEE Conference on Industrial Electronics and Applications, pp. 763–768 (2007)Google Scholar
  8. 8.
    Yin, P.Y.: Ant colony search algorithms for optimal polygonal approximation of plane curves. Pattern Recognition 36, 1783–1797 (2003)MATHCrossRefGoogle Scholar
  9. 9.
    Dong, F., Xiao, R., Zhong, Y.: An approach to polygonal approximation of digital curves based on discrete particle swarm algorithm. Journal of Universal Computer Science 13(10), 1449–1461 (2007)Google Scholar
  10. 10.
    Yin, P.Y.: Genetic particle swarm optimization for polygonal approximation of digital curves. Image Processing, Analysis, Recognition and Understanding 16(2), 223–233 (2006)CrossRefGoogle Scholar
  11. 11.
    Yin, P.Y.: Vision Systems: Segmentation and Pattern Recognition. Polygonal Approximation of Digital Curves Using the State-of-the-art Metaheuristics, 451–464 (2007)Google Scholar
  12. 12.
    Fogel, L.J.: Intelligence through simulated evolution: forty years of evolutionary programming. John Wiley & Sons (1999)Google Scholar
  13. 13.
    Alvarado-Velazco, P.B., Ayala-Ramirez, V., Sanchez-Yanez, R.E.: Polygonal approximation of digital curves using evolutionary programming. Acta Universitaria 22(NE-1), 12–20 (2012)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Paola Berenice Alvarado-Velazco
    • 1
  • Victor Ayala-Ramirez
    • 1
  1. 1.Electronics Engineering DepartmentUniversidad de Guanajuato DICISSalamancaMexico

Personalised recommendations