Abstract
Total variation denoising (TVD) algorithm is suitable for the restoration of piecewise constant signals (PCS). The amplitude accuracy of discontinuities affects the overall denoising effect. Improving the amplitude accuracy of discontinuities is the difficulty of PCS denoising. Recent papers have presented TVD-based algorithms to improve the denoising precision. After applying the forward–backward splitting (FBS) method, the iteration of the denoising is comprised by the traditional TVD. This paper proposes a new upper bound function to enhance the accuracy of the traditional TVD method. Then, the enhanced TVD is used to update the TVD-based algorithm to improve the denoising efficiency. The experimental results demonstrate that the proposed method has superior performance compared to other methods in PCS denoising.
Similar content being viewed by others
Abbreviations
- N :
-
The number of sampling points of the signal
- t, r :
-
The tth and rth iterations, \(t\in \mathbb {N^*}\), and \(r\in \mathbb {N^*}\)
- \(\varvec{y}\) :
-
Noisy signal, \(\varvec{y}\in \mathbb {R}^N\)
- \(\varvec{x}\) :
-
Denoised signal, \(\varvec{x}\in \mathbb {R}^N\)
- \(\varvec{x}_{r}\) :
-
TVD result of the rth iteration, \(\varvec{x}_r\in \mathbb {R}^N\)
- \(\hat{\varvec{x}}_{r}\) :
-
Tolerance error TVD result of the rth iteration, \(\hat{\varvec{x}}_r\in \mathbb {R}^N\)
- \(\Vert \varvec{x}\Vert _1\) :
-
The \(\ell _1\) norm of \(\varvec{x}\)
- \(\Vert \varvec{x}\Vert _2\) :
-
The \(\ell _2\) norm of \(\varvec{x}\)
- \(\varvec{D}\) :
-
First-order difference matrix
- x(n):
-
The nth component of \(\varvec{x}\) and \(n=1,2,\ldots ,N\)
- \(\eta\) :
-
Penalty, \(\eta :\mathbb {R}^N\rightarrow \mathbb {R}\)
- \(\lambda\) :
-
Regularization parameter
- \(G_{\lambda \eta (\cdot )}\) :
-
Objective function, \(G_{\lambda \eta (\cdot )}: \mathbb {R}^N\rightarrow \mathbb {R}\)
- \(\varvec{x}_{\lambda \eta (\cdot )}^*\) :
-
Minimizer of \(G_{\lambda \eta (\cdot )}\), \(\varvec{x}_{\lambda \eta (\cdot )}^*\in \mathbb {R}^N\)
- \(\mathrm{TVD}(\varvec{y};\lambda )\) :
-
Result of TVD for the noisy signal \(\varvec{y}\)
- \(\mathrm{eTVD}(\varvec{y};\lambda )\) :
-
Result of the enhanced TVD for the noisy signal \(\varvec{y}\)
- \(j_t, h_r, H_r\) :
-
Upper bound functions, \(j_t:\mathbb {R}\rightarrow \mathbb {R}\), \(h_r:\mathbb {R}\rightarrow \mathbb {R}\), and \(H_r:\mathbb {R}^N\rightarrow \mathbb {R}\)
References
Little MA, Jones NS (2011) Generalized methods and solvers for noise removal from piecewise constant signals. i. background theory. Proc R Soc A 467(2135):3088–3114. https://doi.org/10.1098/rspa.2010.0671
Yi C, Lv Y, Dang Z, Xiao H (2016) A novel mechanical fault diagnosis scheme based on the convex 1-d second-order total variation denoising algorithm. Appl Sci 6(12):403. https://doi.org/10.1007/s10851-019-00937-5
Selesnick I, Lanza A, Morigi S, Sgallari F (2020) Non-convex total variation regularization for convex denoising of signals. J Math Imaging Vis 62:825–841. https://doi.org/10.1007/s10851-019-00937-5
Prateek GV, Ju YE, Nehorai A (2021) Sparsity-assisted signal denoising and pattern recognition in time-series data. Circ Syst Signal Process. https://doi.org/10.1007/s00034-021-01774-x
Chambolle A, Duval V, Peyré G, Poon C (2016) Geometric properties of solutions to the total variation denoising problem. Inv Probl 33(1):015002. https://doi.org/10.1088/0266-5611/33/1/015002
Ozmen G, Ozsen S (2018) A new denoising method for fMRI based on weighted three-dimensional wavelet transform. Neural Comput Appl 29(8):263–276. https://doi.org/10.1007/s00521-017-2995-7
Bayer FM, Kozakevicius AJ, Cintra RJ (2019) An iterative wavelet threshold for signal denoising. Signal Process 162:10–20. https://doi.org/10.1016/j.sigpro.2019.04.005
Hyeokho C, Richard GB (2003) Interpolation and denoising of piecewise smooth signals by wavelet regularization. In: Michael AU, Akram A, Andrew FL (eds) Wavelets: applications in signal and image processing X, vol 5207. SPIE, Bellingham, pp 16–27
Weinmann A, Storath M, Demaret L (2015) The \(\ell _1\)-potts functional for robust jump-sparse reconstruction. SIAM J Numer Anal 53(1):644–672. https://doi.org/10.1137/120896256
Storath M, Weinmann A, Demaret L (2014) Jump-sparse and sparse recovery using potts functionals. IEEE Trans Signal Process 62(14):3654–3666. https://doi.org/10.1109/TSP.2014.2329263
Zhang H, Wu C, Zhang J, Deng J (2015) Variational mesh denoising using total variation and piecewise constant function space. IEEE Trans Vis Comput Graph 21(7):873–886. https://doi.org/10.1109/TVCG.2015.2398432
Du H, Liu Y (2018) Minmax-concave total variation denoising. Signal Image Video P 12(6):1027–1034. https://doi.org/10.1007/s11760-018-1248-2
Fang Y, Ma Z, Zheng H, Ji W (2020) Trainable \(\rm {TV}\)-\(l^1\) model as recurrent nets for low-level vision. Neural Comput Appl 32(18):14603–14611. https://doi.org/10.1007/s00521-020-05146-5
Lv D, Cao W, Hu W, Wu M (2021) A new total variation denoising algorithm for piecewise constant signals based on non-convex penalty. In: Zhang H, Yang Z, Zhang Z, Wu Z, Hao T (eds) Neural computing for advanced applications, vol 1449. Springer, Singapore, pp 633–644
Selesnick I, Farshchian M (2017) Sparse signal approximation via nonseparable regularization. IEEE Trans Signal Process 65(10):2561–2575. https://doi.org/10.1109/TSP.2017.2669904
Selesnick I, Parekh A, Bayram I (2015) Convex 1-d total variation denoising with non-convex regularization. IEEE Signal Process Lett 22(2):141–144. https://doi.org/10.1109/LSP.2014.2349356
Selesnick I (2017) Total variation denoising via the moreau envelope. IEEE Signal Process Lett 24(2):216–220. https://doi.org/10.1109/LSP.2017.2647948
Combettes PL, Pesquet JC (2011) Proximal splitting methods in signal processing. In: Bauschke H, Burachik R, Combettes P, Elser V, Luke D, Wolkowicz H (eds) Fixed-Point algorithms for inverse problems in science and engineering, vol 49. Springer optimization and its applications. Springer, New York, pp 185–212
Figueiredo TMA, Dias JB, Oliveira JP, Nowak RD (2008) On total variation denoising: a new majorization-minimization algorithm and an experimental comparisonwith wavalet denoising. Paper presented at the international conference on image processing, Atlanta, GA, USA, 8–11 Oct 2006. https://doi.org/10.1109/ICIP.2006.313050
Zhang Y, Kang R, Peng X, Wang J, Zhu J, Peng J, Liu H (2020) Image denoising via structure-constrained low-rank approximation. Neural Comput Appl 32(16):12575–12590. https://doi.org/10.1007/s00521-020-04717-w
Iglesias JA, Mercier G, Scherzer O (2018) A note on convergence of solutions of total variation regularized linear inverse problems. Inv Probl 34(5):055011. https://doi.org/10.1088/1361-6420/aab92a
Acknowledgements
This work was supported by the Hubei Province Natural Science Foundation under Grant 2020CFA031, the Inner Mongolia Natural Science Foundation under Grant 2019BS06004, the National Natural Science Foundation of China under Grants 61773354, 61903345 and 62003318, and the Fundamental Research Funds for the Central Universities under Grant 137-162301202638.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest to this work.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Lv, D., Cao, W., Hu, W. et al. Denoising of piecewise constant signal based on total variation. Neural Comput & Applic 34, 16341–16349 (2022). https://doi.org/10.1007/s00521-022-06937-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-022-06937-8