Image denoising using learned dictionaries and K-SVD

  • Abstract
  • Keywords
  • References
  • Untitled
  • PDF
  • Abstract

    The zero-mean white and homogeneous Gaussian additive unwanted signal must be deleted from the known unique image whenever we are going to tackle the image denoising issue. Here the approach that has well thought-out is relying on the sparse and unnecessary representations over trained dictionaries. The valuable image content which depicts in a dictionary is completed by K-SVD algorithm. Here we take two options of training from the tarnished image itself or training on an amount of pure good quality image database. As we know that the K-SVD is confined in managing very small image patches that are extendable in deploy an arbitrary image sizes by defining a global image prior that pressurizes sparsity over patches in all spots of the image. Such a straightforward and efficient denoising algorithm is done by Bayesian treatment. This makes the paper very effectively surpassing all the up to date published papers on image denoising and the situation of art denoising concert is improvised.



  • Keywords

    Image Denoising; K-SVD; Dictionaries.

  • References

      [1] L. Donoho and I. M. Johnstone, “Minimax estimation via wavelet shrinkage,” Ann. Statist., vol. 26, no. 3, pp. 879–921, Jun. 1998.

      [2] E. P. Simoncelli and E. H. Adelson, “Noise removal via Bayesian wavelet coring,” in Proc. Int. Conf. Image Processing, Laussanne, Switzerland, Sep. 1996.

      [3] A. Chambolle, R. A. DeVore, N.-Y. Lee, and B. J. Lucier, “Nonlinear wavelet image processing: Variational problems, compression, and noise removal through wavelet shrinkage,” IEEE Trans. Image Process., vol. 7, no. 3, pp. 319–335, Mar. 1998.

      [4] M. N. Do and M. Vetterli, Contourlets, Beyond Wavelets, G. V. Welland, Ed. New York: Academic, 2003. [14] M. N. Do and M. Vetterli, “Framing pyramids,” IEEE Trans. Signal Process., vol. 51, pp. 2329–2342, Sep. 2003.

      [5] S. Mallat and E. LePennec, “Sparse geometric image representationwith bandelets,” IEEE Trans. Image Process., vol. 14, no. 4, pp. 423–438, Apr. 2005.





Article ID: 18062
DOI: 10.14419/ijet.v7i4.18062

Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.