Bayesian Non-local Means Filter, Image Redundancy and Adaptive Dictionaries for Noise Removal

  • Charles Kervrann
  • Jérôme Boulanger
  • Pierrick Coupé
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4485)

Abstract

Partial Differential equations (PDE), wavelets-based methods and neighborhood filters were proposed as locally adaptive machines for noise removal. Recently, Buades, Coll and Morel proposed the Non-Local (NL-) means filter for image denoising. This method replaces a noisy pixel by the weighted average of other image pixels with weights reflecting the similarity between local neighborhoods of the pixel being processed and the other pixels. The NL-means filter was proposed as an intuitive neighborhood filter but theoretical connections to diffusion and non-parametric estimation approaches are also given by the authors. In this paper we propose another bridge, and show that the NL-means filter also emerges from the Bayesian approach with new arguments. Based on this observation, we show how the performance of this filter can be significantly improved by introducing adaptive local dictionaries and a new statistical distance measure to compare patches. The new Bayesian NL-means filter is better parametrized and the amount of smoothing is directly determined by the noise variance (estimated from image data) given the patch size. Experimental results are given for real images with artificial Gaussian noise added, and for images with real image-dependent noise.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Awate, S.P., Whitaker, R.T.: Higher-order image statistics for unsupervised, information-theoretic, adaptive, image filtering. In: CVPR’05, San Diego (2005)Google Scholar
  2. 2.
    Guichard, F., Paragios, N., Azzabou, N.: Random Walks, Constrained Multiple Hypothesis Testing and Image Enhancement. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3951, pp. 379–390. Springer, Heidelberg (2006)Google Scholar
  3. 3.
    Barash, D., Comaniciu, D.: A Common framework for nonlinear diffusion, adaptive smoothing, bilateral filtering and mean shift. Image Vis. Comp. 22, 73–81 (2004)CrossRefGoogle Scholar
  4. 4.
    Buades, A., Coll, B., Morel, J.M.: A review of image denoising methods, with a new one. Multiscale Modeling and Simulation 4, 490–530 (2005)MATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    Buades, A., Coll, B., Morel, J.M.: The staircasing effect in neighborhood filters and its solution. IEEE T. Image Process. 15 (2006)Google Scholar
  6. 6.
    Bunea, F., Nobel, A.B.: Sequential procedures for aggregating arbitrary estimators of a conditional mean (under revision) (2005)Google Scholar
  7. 7.
    Coupé, P., Yger, P., Barillot, C.: Fast non-local means denoising for 3D MR images. In: Larsen, R., Nielsen, M., Sporring, J. (eds.) MICCAI 2006. LNCS, vol. 4190, Springer, Heidelberg (2006)Google Scholar
  8. 8.
    Criminisi, A., Pérez, P., Toyama, K.: Region filling and object removal by exemplar-based inpainting. IEEE T. Image Process. 13, 1200–1212 (2004)CrossRefGoogle Scholar
  9. 9.
    De Bonet, J.S.: Noise reduction through detection of signal redundancy. Rethinking Artificial Intelligence, MIT AI Lab. (1997)Google Scholar
  10. 10.
    Dabov, K., et al.: Image denoising with block-matching and 3D filtering. In: Electronic Imaging’06, San Jose, California USA. Proc. of SPIE, vol. 6064, no. 6064A-30 (2006)Google Scholar
  11. 11.
    Efros, A., Leung, T.: Texture synthesis by non-parametric sampling. In: ICCV’99, Kerkyra (1999)Google Scholar
  12. 12.
    Elad, M.: On the bilateral filter and ways to improve it. IEEE T. Image Process. 11, 1141–1151 (2002)CrossRefMathSciNetGoogle Scholar
  13. 13.
    Elad, M., Aharon, M.: Image denoising via learned dictionaries and sparse representation. In: CVPR’06, New York (2006)Google Scholar
  14. 14.
    Faraji, H., MacLean, J.W.: CCD noise removal in digital images. IEEE T. Image Process. 15(9), 2676–2685 (2006)CrossRefGoogle Scholar
  15. 15.
    Geman, D., et al.: Boundary detection by constrained optimization. IEEE T. Patt. Anal. Mach. Intell. 12, 609–628 (1990)CrossRefGoogle Scholar
  16. 16.
    Gilboa, G., Osher, S.: Nonlocal linear image regularization and supervised segmentation. UCLA CAM Report 06-47 (2006)Google Scholar
  17. 17.
    Godtliebsen, F., Spjotvoll, E., Marron, J.S.: A nonlinear Gaussian filter applied to images with discontinuities. J. Nonparametric Stat. 8, 21–43 (1997)MATHCrossRefMathSciNetGoogle Scholar
  18. 18.
    Hirakawa, K., Parks, T.W.: Image denoising using total least squares. IEEE T. Image Process. 15(9), 2730–2742 (2006)CrossRefGoogle Scholar
  19. 19.
    Katkovnik, V., Egiazarian, K., Astola, J.: Adaptive window size image denoising based on intersection of confidence intervals (ICI) rule. J. Math. Imag. Vis. 16, 223–235 (2002)MATHCrossRefMathSciNetGoogle Scholar
  20. 20.
    Kervrann, C., Heitz, F.: A Markov random field model-based approach to unsupervised texture segmentation using local and global spatial statistics. IEEE T. Image Process. 4, 856–862 (1995)CrossRefGoogle Scholar
  21. 21.
    Kervrann, C., Boulanger, J.: Unsupervised patch-based image regularization and representation. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3951, Springer, Heidelberg (2006)Google Scholar
  22. 22.
    Kinderman, S., Osher, S., Jones, P.W.: Deblurring and denoising of images by nonlocal functionals. Multiscale Modeling and Simulation 4, 1091–1115 (2005)CrossRefMathSciNetGoogle Scholar
  23. 23.
    Lee, J.S.: Digital image smoothing and the sigma filter. Comp. Vis. Graph. Image Process. 24, 255–269 (1983)CrossRefGoogle Scholar
  24. 24.
    Loupas, T., McDicken, W.N., Allan, P.L.: An adaptive weighted median filter for speckle suppression inmedical ultrasonic images. IEEE T. Circ. Syst. 36, 129–135 (1989)CrossRefGoogle Scholar
  25. 25.
    Lukin, A.: A multiresolution approach for improving quality of image denoising algorithms. In: ICASSP’06, Toulouse (2006)Google Scholar
  26. 26.
    Mahmoudi, M., Sapiro, G.: Fast image and video denoising via nonlocal means of similar neighborhoods. IEEE Signal Processing Letters 12(12), 839–842 (2005)CrossRefGoogle Scholar
  27. 27.
    Mrazek, P., Weickert, J., Bruhn, A.: On robust estimation and smoothing with spatial and tonal kernels. Preprint 51, U. Bremen (2004)Google Scholar
  28. 28.
    Mumford, D., Shah, J.: Optimal approximations by piecewise smooth functions and variational problems. Comm. Pure and Appl. Math. 42, 577–685 (1989)MATHCrossRefMathSciNetGoogle Scholar
  29. 29.
    Perona, P., Malik, J.: Scale space and edge detection using anisotropic diffusion. IEEE T. Patt. Anal. Mach. Intell. 12, 629–639 (1990)CrossRefGoogle Scholar
  30. 30.
    Polzehl, J., Spokoiny, V.: Adaptive weights smoothing with application to image restoration. J. Roy. Stat. Soc. B 62, 335–354 (2000)CrossRefMathSciNetGoogle Scholar
  31. 31.
    Portilla, J., et al.: Image denoising using scale mixtures of Gaussians in the wavelet domain. IEEE T. Image Process. 12, 1338–1351 (2003)CrossRefMathSciNetGoogle Scholar
  32. 32.
    Roth, S., Black, M.J.: Fields of experts: a framework for learning image priors with applications. In: CVPR’05, San Diego (2005)Google Scholar
  33. 33.
    Rudin, L., Osher, S., Fatemi, E.: Nonlinear Total Variation based noise removal algorithms. Physica D 60(1-4), 259–268 (1992)MATHCrossRefGoogle Scholar
  34. 34.
    Tomasi, C., Manduchi, R.: Bilateral filtering for gray and color images. In: ICCV’98, Bombay (1998)Google Scholar
  35. 35.
    Tschumperlé, D.: Curvature-preserving regularization of multi-valued images using PDE’s. In: Leonardis, A., Bischof, H., Pinz, A. (eds.) ECCV 2006. LNCS, vol. 3951, Springer, Heidelberg (2006)Google Scholar
  36. 36.
    Wang, Z., Zhang, D.: Restoration of impulse noise corrupted images using long-range correlation. IEEE Signal Processing Letters 5, 4–6 (1998)CrossRefGoogle Scholar
  37. 37.
    Weickert, J.: Anisotropic Diffusion in Image Processing. Teubner-Verlag, Stuttgart (1998)MATHGoogle Scholar
  38. 38.
    van de Weijer, J., van den Boomgaard, R.: Local mode filtering. In: CVPR’01, Kauai (2001)Google Scholar
  39. 39.
    Zhang, D., Wang, Z.: Image information restoration based on long-range correlation. IEEE T. Circ. Syst. Video Technol. 12, 331–341 (2002)CrossRefGoogle Scholar

Copyright information

© Springer Berlin Heidelberg 2007

Authors and Affiliations

  • Charles Kervrann
    • 1
    • 3
  • Jérôme Boulanger
    • 1
    • 3
  • Pierrick Coupé
    • 2
  1. 1.INRIA, IRISA, Campus de Beaulieu, 35 042 RennesFrance
  2. 2.Université de Rennes 1, IRISA, Campus de Beaulieu, 35 042 RennesFrance
  3. 3.INRA - MIA, Domaine de Vilvert, 78 352 Jouy-en-JosasFrance

Personalised recommendations