Research of Image Segmentation Base on PCNN Method

Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 217)

Abstract

The space incoherent and small changes in the amplitude of input image can be compensated by utilizing the character of PCNN that the similar input neurons can pulse simultaneously. And then the small regions of the image can be segmented very well by means of adjusting a threshold parameter of PCNN, therefore, the complete information of the image can be preserved and the quality of the image can be improved too. It is more effective to enhance the segmentation quality of the image than the method that enhanced an image through the image enhanced function DECORRSTRETCH of MATLAB and carrying out the information of spectral bands which will be normalized, finally adjusting the threshold.

Keywords

PCNN Image segmentation Small regions Threshold adjustment 

Notes

Acknowledgments

The authors are greatly indebted to anonymous referees for their constructive comments. The work described in this paper was partially supported by the Natural Science Foundation Project of CQ CSTC (Grant No. CSTC2009BB6388), and Applying Basic Research Program of Chongqing Education Committee (No. KJ110628, KJ100611, KJ110617) and Excellent Talents Project of Chongqing Education Committee.

References

  1. 1.
    Johnson JL, Padgett ML (1999) PCNN models and applications. IEEE Trans Neural Networks (S1045-9227) 10(3):480–498Google Scholar
  2. 2.
    Ranganath HS, Kuntimad G (1999) Object detection using pulse coupled neural networks. IEEE Trans Neural Networks 10(3):615–620CrossRefGoogle Scholar
  3. 3.
    Xiaodong G, Haiming W, Daoheng Y (2001) Binary image restoration using pulse coupled neural network. 8th international conference on neural information processing, vol 5 ICONIP-2001, pp 25–26Google Scholar
  4. 4.
    Jcaufield H, Kinser JM (1999) Finding shortest path in the shortest time using PCNN. IEEE Trans Neural Networks 10(3):604–606CrossRefGoogle Scholar
  5. 5.
    Wang DL (2000) On connectedness: a solution based on oscillatory correlation. IEEE Trans Neural Network 12(1):181–194Google Scholar
  6. 6.
    Ma Y, Zhan K, Wang Z (2010) Applications of pulse-coupled neural networks, vol 67. Higher Education Press, Beijing, pp 3–4Google Scholar
  7. 7.
    Zhang Z, Ma G (2009) PCNN model parameter optimization and multi-threshold image segmentation. J Harbin Inst Tech (3)Google Scholar

Copyright information

© Springer-Verlag London 2013

Authors and Affiliations

  1. 1.Department of MathematicsChongqing Normal UniversityChongqingChina
  2. 2.Departments of Computer and Information ScienceChongqing Normal UniversityChongqingChina

Personalised recommendations