Advertisement

Signal, Image and Video Processing

, Volume 12, Issue 6, pp 1027–1034 | Cite as

Minmax-concave total variation denoising

  • Huiqian Du
  • Yilin Liu
Original Paper
  • 129 Downloads

Abstract

Total variation (TV) denoising is a commonly used method for recovering 1-D signal or 2-D image from additive white Gaussian noise observation. In this paper, we define the Moreau enhanced function of \(L_1\) norm as \({\varPhi }_\alpha (x)\) and introduce the minmax-concave TV (MCTV) in the form of \({\varPhi }_\alpha (Dx)\), where D is the finite difference operator. We present that MCTV approaches \(\Vert Dx\Vert _0\) if the non-convexity parameter \(\alpha \) is chosen properly and apply it to denoising problem. MCTV can strongly induce the signal sparsity in gradient domain, and moreover, its form allows us to develop corresponding fast optimization algorithms. We also prove that although this regularization term is non-convex, the cost function can maintain convexity by specifying \(\alpha \) in a proper range. Experimental results demonstrate the effectiveness of MCTV for both 1-D signal and 2-D image denoising.

Keywords

Total variation Signal denoising Non-convex regularization \(L_1\) norm 

References

  1. 1.
    Beck, A., Teboulle, M.: A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J. Imaging Sci. 2(1), 183–202 (2009).  https://doi.org/10.1137/080716542 MathSciNetCrossRefMATHGoogle Scholar
  2. 2.
    Bhattacharya, S., Venkatsh, K.S., Gupta, S.: Background estimation and motion saliency detection using total variation-based video decomposition. Signal Image Video Process. 11(1), 113–121 (2016).  https://doi.org/10.1007/s11760-016-0909-2 CrossRefGoogle Scholar
  3. 3.
    Boyd, S., Parikh, N., Chu, E., Peleato, B., Eckstein, J.: Distributed optimization and statistical learning via the alternating direction method of multipliers. Found. Trends® Mach. Learn. 3(1), 1–122 (2010).  https://doi.org/10.1561/2200000016 CrossRefMATHGoogle Scholar
  4. 4.
    Bresson, X.: A short note for nonlocal tv minimization. Technical report (2009)Google Scholar
  5. 5.
    Chartrand, R.: Shrinkage mappings and their induced penalty functions. In: 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE (2014).  https://doi.org/10.1109/icassp.2014.6853752
  6. 6.
    Chen, L., Gu, Y.: The convergence guarantees of a non-convex approach for sparse recovery. IEEE Trans. Signal Process. 62(15), 3754–3767 (2014).  https://doi.org/10.1109/tsp.2014.2330349 MathSciNetCrossRefGoogle Scholar
  7. 7.
    Combettes, P.L., Pesquet, J.C.: Proximal Splitting Methods in Signal Processing. Springer Optimization and Its Applications, pp. 185–212. Springer, New York (2011).  https://doi.org/10.1007/978-1-4419-9569-8_10 MATHGoogle Scholar
  8. 8.
    Condat, L.: A direct algorithm for 1-d total variation denoising. IEEE Signal Process. Lett. 20(11), 1054–1057 (2013).  https://doi.org/10.1109/lsp.2013.2278339 CrossRefGoogle Scholar
  9. 9.
    Daubechies, I., Defrise, M., Mol, C.D.: An iterative thresholding algorithm for linear inverse problems with a sparsity constraint. Commun. Pure Appl. Math. 57(11), 1413–1457 (2004).  https://doi.org/10.1002/cpa.20042 MathSciNetCrossRefMATHGoogle Scholar
  10. 10.
    Goldstein, T., Osher, S.: The split Bregman method for L1-regularized problems. SIAM J. Imaging Sci. 2(2), 323–343 (2009).  https://doi.org/10.1137/080725891 MathSciNetCrossRefMATHGoogle Scholar
  11. 11.
    Kim, H., Chen, J., Wang, A., Chuang, C., Held, M., Pouliot, J.: Non-local total-variation (NLTV) minimization combined with reweighted L1-norm for compressed sensing CT reconstruction. Phys. Med. Biol. 61(18), 6878–6891 (2016).  https://doi.org/10.1088/0031-9155/61/18/6878 CrossRefGoogle Scholar
  12. 12.
    Knoll, F., Bredies, K., Pock, T., Stollberger, R.: Second order total generalized variation (TGV) for MRI. Magn. Reson. Med. 65(2), 480–491 (2010).  https://doi.org/10.1002/mrm.22595 CrossRefGoogle Scholar
  13. 13.
    Lanza, A., Morigi, S., Sgallari, F.: Convex Image Denoising via Non-convex Regularization. Lecture Notes in Computer Science, pp. 666–677. Springer, New York (2015).  https://doi.org/10.1007/978-3-319-18461-6_53 Google Scholar
  14. 14.
    Lanza, A., Morigi, S., Sgallari, F.: Convex image denoising via non-convex regularization with parameter selection. J. Math. Imaging Vis. 56(2), 195–220 (2016).  https://doi.org/10.1007/s10851-016-0655-7 MathSciNetCrossRefMATHGoogle Scholar
  15. 15.
    Moreau, J.: Inf-convolution des fonctions numriques sur un espace vectoriel. Comptes Rendus de l’Acadmie des Sciences 256, 5047–5049 (1963)MATHGoogle Scholar
  16. 16.
    Nikolova, M.: Analysis of the recovery of edges in images and signals by minimizing nonconvex regularized least-squares. Multiscale Model. Simul. 4(3), 960–991 (2005).  https://doi.org/10.1137/040619582 MathSciNetCrossRefMATHGoogle Scholar
  17. 17.
    Nikolova, M.: Analytical bounds on the minimizers of (nonconvex) regularized least-squares. Inverse Probl. Imaging 2(1), 133–149 (2008).  https://doi.org/10.3934/ipi.2008.2.133 MathSciNetCrossRefMATHGoogle Scholar
  18. 18.
    Rudin, L.I., Osher, S., Fatemi, E.: Nonlinear total variation based noise removal algorithms. Physica D: Nonlinear Phenomena 60(1–4), 259–268 (1992).  https://doi.org/10.1016/0167-2789(92)90242-f MathSciNetCrossRefMATHGoogle Scholar
  19. 19.
    Selesnick, I.: Sparse regularization via convex analysis. IEEE Trans. Signal Process. 65(17), 4481–4494 (2017).  https://doi.org/10.1109/tsp.2017.2711501
  20. 20.
    Selesnick, I.: Total variation denoising via the Moreau envelope. IEEE Signal Process. Lett. 24(2), 216–220 (2017).  https://doi.org/10.1109/lsp.2017.2647948 MathSciNetCrossRefGoogle Scholar
  21. 21.
    Selesnick, I., Farshchian, M.: Sparse signal approximation via nonseparable regularization. IEEE Trans. Signal Process. 65(10), 2561–2575 (2017).  https://doi.org/10.1109/tsp.2017.2669904 MathSciNetCrossRefGoogle Scholar
  22. 22.
    Selesnick, I., Parekh, A., Bayram, I.: Convex 1-d total variation denoising with non-convex regularization. IEEE Signal Process. Lett. 22(2), 141–144 (2015).  https://doi.org/10.1109/lsp.2014.2349356 CrossRefGoogle Scholar
  23. 23.
    Tofighi, M., Kose, K., Cetin, A.E.: Denoising images corrupted by impulsive noise using projections onto the epigraph set of the total variation function (PES-TV). Signal Image Video Process. 9(S1), 41–48 (2015).  https://doi.org/10.1007/s11760-015-0827-8 CrossRefGoogle Scholar
  24. 24.
    Xie, W.S., Yang, Y.F., Zhou, B.: An ADMM algorithm for second-order TV-based MR image reconstruction. Numer. Algorithms 67(4), 827–843 (2014).  https://doi.org/10.1007/s11075-014-9826-z MathSciNetCrossRefMATHGoogle Scholar
  25. 25.
    Xu, J., Feng, X., Hao, Y., Han, Y.: Image decomposition using adaptive second-order total generalized variation. Signal Image Video Process. 8(1), 39–47 (2012).  https://doi.org/10.1007/s11760-012-0420-3 CrossRefGoogle Scholar
  26. 26.
    Yang, S., Wang, J., Fan, W., Zhang, X., Wonka, P., Ye, J.: Analysis of the recovery of edges in images and signals by minimizing nonconvex regularized least-squares. In: Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (2013).  https://doi.org/10.1145/2487575.2487586
  27. 27.
    Zhang, C.H.: Nearly unbiased variable selection under minimax concave penalty. Ann. Stat. 38(2), 894–942 (2010).  https://doi.org/10.1214/09-aos729 MathSciNetCrossRefMATHGoogle Scholar
  28. 28.
    Zhang, X., Burger, M., Bresson, X., Osher, S.: Bregmanized nonlocal regularization for deconvolution and sparse reconstruction. SIAM J. Imaging Sci. 3(3), 253–276 (2010).  https://doi.org/10.1137/090746379 MathSciNetCrossRefMATHGoogle Scholar

Copyright information

© Springer-Verlag London Ltd., part of Springer Nature 2018

Authors and Affiliations

  1. 1.School of Information and ElectronicsBeijing Institute of TechnologyBeijingChina

Personalised recommendations