Learning Stable Nonlinear Cross-Diffusion Models for Image Restoration

  • Sílvia BarbeiroEmail author
  • Diogo Lobo


In this paper, we focus on learning optimized partial differential equation (PDE) models for image filtering. In our approach, the gray-scale images are represented by a vector field of two real-valued functions and the image restoration problem is modeled by an evolutionary process such that the restored image at any time satisfies an initial-boundary value problem of cross-diffusion with reaction type. The coupled evolution of the two components of the image is determined by a nondiagonal matrix that depends on those components. A critical question when designing a good performing filter lies in the selection of the optimal coefficients and influence functions which define the cross-diffusion matrix. We propose to take a PDE based on a nonlinear cross-diffusion process and turn it into a learnable architecture in order to optimize the parameters of the model. In particular, we use a back-propagation technique in order to minimize a cost function related to the quality of the denoising process, while we ensure stability during the learning procedure. Consequently, we obtain improved image restoration models with solid mathematical foundations. The learning framework and resulting models are presented along with related numerical experiments and image comparisons. Making use of synthetic data, the numerical results show the advantages of the proposed methodology by achieving significant improvements.


Nonlinear cross-diffusion Image denoising Machine learning Back-propagation Optimal parameters Stability 



  1. 1.
    Araújo, A., Barbeiro, S., Cuesta, E., Durán, A.: Cross-diffusion systems for image processing: I. The linear case. J. Math. Imaging Vis. 58, 447–467 (2017)MathSciNetCrossRefGoogle Scholar
  2. 2.
    Araújo, A., Barbeiro, S., Cuesta, E., Durán, A.: Cross-diffusion systems for image processing: II. The nonlinear case. J. Math. Imaging Vis. 58, 427–446 (2017)MathSciNetCrossRefGoogle Scholar
  3. 3.
    Araújo, A., Barbeiro, S., Cuesta, E., Durán, A.: A discrete cross-diffusion model for image restoration. In: Quintela, P., Barral, P., Gómez, D., Pena, F.J., Rodríguez, J., Salgado, P., Vázquez-Méndez, M.E. (eds.) Progress in Industrial Mathematics at ECMI 2016, Springer, Cham, pp. 401–408 (2017)CrossRefGoogle Scholar
  4. 4.
    Araújo, A., Barbeiro, S., Serranho, P.: Stability of finite difference schemes for complex diffusion processes. SIAM J. Numer. Anal. 50, 1284–1296 (2012)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Araújo, A., Barbeiro, S., Serranho, P.: Stability of finite difference schemes for nonlinear complex reaction–diffusion processes. IMA J. Numer. Anal. 35, 1381–1401 (2015)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Bernardes, R., Maduro, C., Serranho, P., Araújo, A., Barbeiro, S., Cunha-Vaz, J.: Improved adaptive complex diffusion despeckling filter. Opt. Express 18, 24048–24059 (2010)CrossRefGoogle Scholar
  7. 7.
    Birgin, E.G., Martínez, J.M.: Practical Augmented Lagrangian Methods for Constrained Optimization, Fundamentals of Algorithms, Society for Industrial and Applied Mathematics (2014)Google Scholar
  8. 8.
    Buhmann, M.D.: Radial Basis Functions: Theory and Implementations. Cambridge University Press, New York (2003)CrossRefGoogle Scholar
  9. 9.
    Chan, T.F., Shen, L.: Stability analysis of difference schemes for variable coefficient schrodinger type equations. SIAM J. Numer. Anal. 24, 336–349 (1987)MathSciNetCrossRefGoogle Scholar
  10. 10.
    Chen, Y., Pock, T.: Trainable nonlinear reaction diffusion: a flexible framework for fast and effective image restoration. IEEE Trans. Pattern Anal. Mach. Intell. 39, 1256–1272 (2016)CrossRefGoogle Scholar
  11. 11.
    Crete, F., Dolmiere, T., Ladret, P., Nicolas, M.: The blur effect: perception and estimation with a new no-reference perceptual blur metric. In: Human Vision and Electronic Imaging (2007)Google Scholar
  12. 12.
    Gilboa, G., Sochen, N., Zeevi, Y.Y.: Image enhancement and denoising by complex diffusion processes. IEEE Trans. Pattern Anal. Mach. Intell. 26, 1020–1036 (2004)CrossRefGoogle Scholar
  13. 13.
    Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016)zbMATHGoogle Scholar
  14. 14.
    Kingma, D., Ba, J.: Adam: a method for stochastic optimization. In: International Conference on Learning Representations (2014)Google Scholar
  15. 15.
    Nordström, N.: Biased anisotropic diffusion: a unified regularization and diffusion approach to edge detection. Image Vis. Comput. 8, 318–327 (1990)CrossRefGoogle Scholar
  16. 16.
    Perona, P., Malik, J.: Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell. 12, 629–639 (1990)CrossRefGoogle Scholar
  17. 17.
    Salinas, H.M., Fernandez, D.C.: Comparison of PDE-based nonlinear diffusion approaches for image enhancement and denoising in optical coherence tomography. IEEE Trans. Med. Imaging 26, 761–771 (2007)CrossRefGoogle Scholar
  18. 18.
    Tsiotsios, C., Petrou, M.: On the choice of the parameters for anisotropic diffusion in image processing. Pattern Recognit. 46, 1369–1381 (2013)CrossRefGoogle Scholar
  19. 19.
    Weickert, J.: A review of nonlinear diffusion filtering. In: ter Haar Romeny, B., Florack, L., Koenderink, J., Viergever, M. (eds.) Scale-Space Theory in Computer Vision. Springer, Berlin, pp. 1–28 (1997)Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.CMUC, Department of MathematicsUniversity of CoimbraCoimbraPortugal

Personalised recommendations