A Self-governing Hybrid Model for Noise Removal

  • Mohammad Reza Hajiaboli
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5414)


Denoising methods based on using fourth-order partial deferential equations (PDEs) are providing a good combination of the noise smoothing and the edge preservation without creating blocky effects on the smooth regions of the image. However, finding an optimal choice of model parameters for numerical solver of these techniques is a challenging problem and generally, these model parameters are image-content dependent. In this paper, a hybrid fourth-order PDE-based filter is proposed so that it does not need a manual adjustment of the model parameters. It is shown that by setting the numerical solver of proposed filter for operation at a minor time step-size derived under a data-independent stability condition, the filter can still provide a significantly fast convergence rate. Therefore, the model parameters are reduced to one parameter estimated by using a well-studied mechanism applying in the second-order nonlinear diffusion denoising techniques. Simulation results show that the proposed method can provide a denoised image with higher quality in comparison with that of the existing methods.


Denoising Diffusion Laplacian Gradient Convergence 


  1. 1.
    Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence 12(7), 629–639 (1990)CrossRefGoogle Scholar
  2. 2.
    Catte, F., et al.: Image selective smoothing and edge detection by nonlinear diffusion. SIAM J. Numer. Anal. 29(1), 182–193 (1992)MathSciNetCrossRefzbMATHGoogle Scholar
  3. 3.
    Black, M.J., et al.: Robust anisotropic diffusion. IEEE Transactions on Image Processing 7(3), 421–432 (1998)MathSciNetCrossRefGoogle Scholar
  4. 4.
    You, Y.L., Kaveh, M.: Fourth-order partial differential equations for noise removal. IEEE Transactions on Image Processing 9(10), 1723–1730 (2000)MathSciNetCrossRefzbMATHGoogle Scholar
  5. 5.
    Lysaker, M., Lundervold, A., Tai, X.-C.: Noise removal using fourth-order partial differential equation with applications to medical magnetic resonance images in space and time. IEEE Tran. on Image Processing 12(12), 1579–1590 (2003)CrossRefzbMATHGoogle Scholar
  6. 6.
    Rajan, J., Kannan, K., Kaimal, M.R.: An Improved hybrid model for molecular image denoising. Journal of Mathematical Imaging and Vision 31, 73–79 (2008)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Li, F., et al.: Image restoration combining a total variational filter and a fourth-order filter. Journal of Visual Communication and Image Representation 18, 322–330 (2007)CrossRefGoogle Scholar
  8. 8.
    You, Y.-L., et al.: Behavioral analysis of anisotropic diffusion in image processing. IEEE Trans. Image Processing 5, 1539–1553 (1996)CrossRefGoogle Scholar
  9. 9.
    Hamza, A.B., et al.: Removing noise and preserving details with relaxed median filters. Journal of Mathematical Imaging and Vision 11(2), 161–177 (1999)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Nixon, M., Aguado, A.: Feature Extraction and Image Processing. Oxford, Newnes (2002)Google Scholar
  11. 11.
    Guido, G., et al.: Nonlinear anisotropic filtering of MRI data. IEEE Transactions on Medical Imaging 11(2), 221–232 (1992)CrossRefGoogle Scholar
  12. 12.
    Canny, J.F.: A computational approach to edge detection. IEEE Transactions on Pattern Analysis and Machine Intelligence 8(6), 679–698 (1986)CrossRefGoogle Scholar
  13. 13.
    Stoer, J., Bulirsch, R.: Introduction to Numerical Analysis. Texts in Applied Mathematics 12. Springer, New York (2002)CrossRefzbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Mohammad Reza Hajiaboli
    • 1
  1. 1.Department of Electrical and Computer EngineeringConcordia UniversityMontrealCanada

Personalised recommendations