Advertisement

Science in China Series F: Information Sciences

, Volume 51, Issue 12, pp 2115–2125 | Cite as

Improved image filter based on SPCNN

  • YuDong Zhang
  • LeNan Wu
Article

Abstract

By extraction of the thoughts of non-linear model and adaptive model match, an improved Nagao filter is brought. Meanwhile a technique based on simplified pulse coupled neural network and used for noise positioning, is put forward. Combining the two methods above, we acquire a new method that can restore images corrupted by salt and pepper noise. Experiments show that this method is more preferable than other popular ones, and still works well while noise density fluctuates severely.

Keywords

Nagao filter pulse coupled neural network image smoothing image de-noising salt and pepper noise edge preserving 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Davis L S, Rosenfeld A. Noise cleaning by iterated local averaging. IEEE Trans Syst, Man and Cybernetics, 1978, (7): 705–710Google Scholar
  2. 2.
    Nodes T A, Gallagher N C Jr. Median filters: some modifications and their properties. IEEE Trans Acoust, Speech, Sig Proc, 1982, 30(5): 739–746CrossRefGoogle Scholar
  3. 3.
    Brownrigg D R K. The weighted median filter. Commun Associat Comput Mach, 1984, 27(8): 807–818Google Scholar
  4. 4.
    Ko S J, Lee Y H. Center weighted median filters and their applications to image enhancement. IEEE Trans Circ Syst, 1991, 38(9): 984–993CrossRefGoogle Scholar
  5. 5.
    Agaian S, Choi D S, Noonan J. Image compression using fuzzy subband decomposition. In: Yeneds J, ed. FUZZ-IEEE 2000-95th IEEE International Conference on Fuzzy Syst, San Antonio, TX, USA: Institute of Electrical and Electronics Engineers Inc, 2004. 894–899Google Scholar
  6. 6.
    Senel H G, Alan P II R, Benoit D. Topological median filters. IEEE Trans Image Proc, 2002, 11(2): 89–104CrossRefGoogle Scholar
  7. 7.
    Sun T, Neuvo Y. Detail-preserving median based filter in image processing. Patt Recog Lett, 1994, 15(4): 341–347CrossRefGoogle Scholar
  8. 8.
    Yuan S Q, Tan Y H. A median-subset-type adaptive median filter. J Imag Graph, 2007, 12(4): 608–612Google Scholar
  9. 9.
    Ranganath H S. Object detection using pulse coupled neural networks. IEEE Trans Neural Networks, 1999, 10(3): 615–620CrossRefGoogle Scholar
  10. 10.
    Makoto N, Takashi M. Edge preserving smoothing. Comput Graph Imag Proc, 1979, (9): 394–407eGoogle Scholar
  11. 11.
    Zhu J H, Yang X, Li J, et al. Texture analysis based detail preserving smoothing filter. J Imag Graph, 2001, 6(11): 1058–1064Google Scholar
  12. 12.
    Eckhorn R, Reitboeck H H, Arndt M, et al. Feature linking via synchronization among distributed assemblies: simulation of results from Cat Cortex. Neural Comput, 1990, 2(3): 293–307CrossRefGoogle Scholar
  13. 13.
    Gu X D, Guo S D, Yu D H. A new approach for noise reducing of image based on PCNN. J Electr Info Tech, 2002, 24(10): 1304–1309Google Scholar
  14. 14.
    Li Y G, Shi M H, Wei Y W. Image Gauss noise filtering based on PCNN. Comput Engin Appl, 2007, 43(1): 65–68Google Scholar
  15. 15.
    Shi M H, Mao J H, Liang Y. Method for filtering image contaminated with strong Gaussian noise. Comput Appl, 2007, 27(7): 637–640Google Scholar
  16. 16.
    Wang W, Li M, Liu G H. New color image filtering algorithm based on PCNN. Comput Engin Design, 2007, 28(14): 3413–3415Google Scholar
  17. 17.
    Ma Y D, Shi F, Li L. Gaussian noise filter based on PCNN. IEEE ICNNSP, 2003Google Scholar
  18. 18.
    Zhang J Y, Lu Z J, Shi L, et al. Filtering images contaminated with pep and salt type noise with pulse-coupled networks. Sci China Ser F-Inf Sci, 2005, 48(3): 322–334CrossRefGoogle Scholar

Copyright information

© Science in China Press and Springer-Verlag GmbH 2008

Authors and Affiliations

  1. 1.School of Information Science & EngineeringSoutheast UniversityNanjingChina

Personalised recommendations