Abstract
This article considers image deblurring in the presence of impulse noise and proposes a new multi-parameter regularization model for image deblurring based on total variation (TV) and wavelet frame (WF), along with an efficient and effective solving algorithm for this restoring model. On the one hand, it is well known that the TV regularization-based Rudin–Osher–Fatemi model is very effective in preserving sharp edges and object boundaries which are generally the most important features to recover. On the other hand, WF-based approaches for image restoration have proven to be very successful in adaptively exploiting the regularity of natural images. By combining TV regularization and WF regularization, a novel multi-parameter regularization model is proposed for deblurring images in the presence of impulse noise. Numerically, the alternative direction method of multiplier (ADMM) with an adaptive scheme for choosing regularization parameters is provided and applied to this multi-parameter regularization model. Moreover, the convergence analysis of the ADMM is shown in the “Appendix.” Furthermore, numerical experiments involving images corrupted by various types of blurring kernels and different levels of noises indicate that the proposed model and algorithm outperform several state-of-the-art approaches in terms of the restoration quality, especially the ability to mitigate staircasing effects while preserving important features in images.
Similar content being viewed by others
References
L. Bar, A. Brook, N. Sochen, N. Kiryati, Deblurring of color images corrupted by impulsive noise. IEEE Trans. Image Process. 16(4), 1101–1111 (2007). doi:10.1109/TIP.2007.891805
L. Bar, N. Sochen, N. Kiryati, Image deblurring in the presence of salt-and-pepper noise, in Scale Space and PDE Methods in Computer Vision, Lecture Notes in Computer Science, ed. R. Kimmel, N. Sochen, J. Weickert (eds.), vol. 3459 (Springer, Berlin, 2005), pp. 107–118. doi:10.1007/11408031_10
L. Bar, N. Sochen, N. Kiryati, Image deblurring in the presence of impulsive noise. Int. J. Comput. Vis. 70(3), 279–298 (2006). doi:10.1007/s11263-006-6468-1
A. Beck, M. Teboulle, A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J. Imaging Sci. 2(1), 183–202 (2009). doi:10.1137/080716542
K. Bredies, T. Pock, B. Wirth, A convex, lower semicontinuous approximation of Euler’s elastica energy. SIAM J. Math. Anal. 47(1), 566–613 (2015). doi:10.1137/130939493
D. Brownrigg, The weighted median filter. Commun. ACM 27(8), 807–818 (1984). doi:10.1145/358198.358222
J. Cai, R. Chan, M. Niklova, Two-phase approach for deblurring images corrupted by impulse plus Gaussian noise. Inverse Probl. Imaging 2(2), 187–204 (2008). doi:10.3934/ipi.2008.2.187
J. Cai, R. Chan, M. Nikolova, Fast two-phase image deblurring under impulse noise. J. Math. Imaging Vis. 36(1), 46–53 (2010). doi:10.1007/s10851-009-0169-7
J. Cai, R. Chan, Z. Shen, Simultaneous cartoon and texture inpainting. Inverse Probl. Imaging 4(3), 379–395 (2010). doi:10.3934/ipi.2010.4.379
J. Cai, B. Dong, S. Osher, Z. Shen, Image restoration: total variation, wavelet frames, and beyond. J. Am. Math. Soc. 25, 1033–1089 (2012). doi:10.1090/S0894-0347-2012-00740-1
J. Cai, Z. Shen, Framelet based deconvolution. J. Comput. Math. 28(3), 289–308 (2010). doi:10.4208/jcm.1001-m1002
A. Chai, Z. Shen, Deconvolution: a wavelet frame approach. Numer. Math. 106(4), 529–587 (2007). doi:10.1007/s00211-007-0075-0
R. Chan, Y. Dong, M. Hintermüller, An efficient two-phase L1-TV method for restoring blurred images with impulse noise. IEEE Trans. Image Process. 19(7), 1731–1739 (2010). doi:10.1109/TIP.2010.2045148
R. Chan, M. Tao, X. Yuan, Constrained total variation deblurring models and fast algorithms based on alternating direction method of multipliers. SIAM J. Imaging Sci. 6(1), 680–697 (2013). doi:10.1137/110860185
T. Chan, S. Esedoglu, Aspects of total variation regularized L1 function approximation. SIAM J. Appl. Math. 65(5), 1817–1837 (2005). doi:10.1137/040604297
T. Chen, H. Wu, Space variant median filters for the restoration of impulse noise corrupted images. IEEE Trans. Circuits Syst. II Analog. Digit. Signal Process. 48(8), 784–789 (2001). doi:10.1109/82.959870
C. Clason, B. Jin, K. Kunisch, A duality-based splitting method for L1-TV image restoration with automatic regularization parameter choice. SIAM J. Sci. Comput. 32(3), 1485–1505 (2010). doi:10.1137/090768217
I. Daubechies, B. Han, A. Ron, Z. Shen, Framelets: MRA-based constructions of wavelet frames. Appl. Comput. Harm. Anal. 14(1), 1–46 (2003). doi:10.1016/S1063-5203(02)00511-0
D. Dobson, F. Santosa, Recovery of blocky images from noisy and blurred data. SIAM J. Appl. Math. 56(4), 1181–1198 (1996). doi:10.1137/S003613999427560X
B. Dong, H. Ji, J. Li, Z. Shen, Y. Xu, Wavelet frame based blind image inpainting. Appl. Comput. Harm. Anal. 32(2), 268–279 (2012). doi:10.1016/j.acha.2011.06.001
B. Dong, Z. Shen, MRA based wavelet frames and applications, in IAS Lecture Notes Series, Summer Program on the Mathematics of Image Processing, Park City Mathematics Institute, vol 19, pp. 7–158 (2010)
Y. Dong, M. Hintermller, M. Neri, An efficient primal-dual method for L1TV image restoration. SIAM J. Imaging Sci. 2(4), 1168–1189 (2009). doi:10.1137/090758490
M. Elad, Sparse and Redundant Representations: From Theory to Applications in Signal and Image Processing (Springer, New York, 2010). doi:10.1007/978-1-4419-7011-4
M. Elad, M. Aharon, Image denoising via sparse and redundant representations over learned dictionaries. IEEE Trans. Image Process. 15(12), 3736–3745 (2006). doi:10.1109/TIP.2006.881969
E. Esser, Applications of Lagrangianbased alternating direction methods and connections to split Bregman. Technical Report, UCLA C.A.M. Report, 09-31 (2009)
S. Foucart, H. Rauhut, A mathematical introduction to compressive sensing. Applied and Numerical Harmonic Analysis. Birkhäuser Basel (2013). doi:10.1007/978-0-8176-4948-7
H. Fu, M. Ng, M. Nikolova, J. Barlow, Efficient minimization methods of mixed l2–l1 and l1–l1 norms for image restoration. SIAM J. Sci. Comput. 27(6), 1881–1902 (2006). doi:10.1137/040615079
D. Gabay, B. Mercier, A dual algorithm for the solution of nonlinear variational problems via finite element approximation. Comput. Math. Appl. 2(1), 17–40 (1976). doi:10.1016/0898-1221(76)90003-1
R. Glowinski, Numerical Methods for Nonlinear Variational Problems (Springer, New York, 1984). doi:10.1007/978-3-662-12613-4
R. Glowinski, A. Marroco, Sur l’approximation, par éléments finis d’ordre un, et la résolution, par pénalisation-dualité d’une classe de problèmes de Dirichlet non linéaires. Rev. Francaise d’Automat. Inf. Recherche Opérationelle, 9 (R2), 41–76 (1975). http://eudml.org/doc/193269
R. Glowinski, P.L. Tallec, Augmented Lagrangian and operator-splitting methods in nonlinear mechanics, chap. Augmented Lagrangian Methods for the Solution of Variational Problems, pp. 45–121. SIAM (1989). doi:10.1137/1.9781611970838.ch3
Z. Gong, Z. Shen, K. Toh, Image restoration with mixed or unknown noises. Multiscale Model. Simul. 12(2), 458–487 (2014). doi:10.1137/130904533
Y. Huang, D. Lu, T. Zeng, Two-step approach for the restoration of images corrupted by multiplicative noise. SIAM J. Sci. Comput. 35(6), 2856–2873 (2013). doi:10.1137/120898693
H. Hwang, R. Haddad, Adaptive median filters: new algorithms and results. IEEE Trans. Image Process. 4(4), 499–502 (1995). doi:10.1109/83.370679
K. Ito, B. Jin, T. Takeuchi, A regularization parameter for nonsmooth Tikhonov regularization. SIAM J. Sci. Comput. 33(3), 1415–1438 (2011). doi:10.1137/100790756
R. Jia, H. Zhao, W. Zhao, Convergence analysis of the Bregman method for the variational model of image denoising. Appl. Comput. Harm. Anal. 27(3), 367–379 (2009). doi:10.1016/j.acha.2009.05.002
S. Ko, Y. Lee, Center weighted median filters and their applications to image enhancements. IEEE Trans. Circuits Syst. 38(9), 984–993 (1991). doi:10.1109/31.83870
Y. Li, L. Shen, D. Dai, B. Suter, Framelet algorithms for de-blurring images corrupted by impulse plus Gaussian noise. IEEE Trans. Image Process. 20(7), 1822–1837 (2011). doi:10.1109/TIP.2010.2103950
L. Ma, J. Yu, T. Zeng, Sparse representation prior and total variation-based image deblurring under impulse noise. SIAM J. Imaging Sci. 6(4), 2258–2284 (2013). doi:10.1137/120866452
J. Mairal, F. Bach, J. Ponce, G. Sapiro, A. Zisserman, Non-local sparse models for image restoration, in IEEE International Conference on Computer Vision (ICCV), pp. 2272–2279 (2009). doi:10.1109/ICCV.2009.5459452
M. Nikolova, Local strong homogeneity of a regularized estimator. SIAM J. Appl. Math. 61(2), 633–658 (2000). doi:10.1137/S0036139997327794
M. Nikolova, Minimizers of cost-functions involving nonsmooth data-fidelity terms. Application to the processing of outliers. SIAM J. Numer. Anal. 40(3), 965–994 (2002). doi:10.1137/S0036142901389165
M. Nikolova, A variational approach to remove outliers and impulse noise. J. Math. Imaging Vis. 20(1–2), 99–120 (2004). doi:10.1023/B:JMIV.0000011326.88682.e5
P. Ochs, Y. Chen, T. Brox, T. Pock, Ipiano: inertial proximal algorithm for nonconvex optimization. SIAM J. Imaging Sci. 7(2), 1388–1419 (2014). doi:10.1137/130942954
D. O’Connor, L. Vandenberghe, Primal-dual decomposition by operator splitting and applications to image deblurring. SIAM J. Imaging Sci 7(3), 1724–1754 (2014). doi:10.1137/13094671X
G. Pok, J. Liu, A. Nair, Selective removal of impulse noise based on homogeneity level information. IEEE Trans. Image Process. 12(1), 85–92 (2003). doi:10.1109/TIP.2002.804278
W. Pratt, Median filtering Technical Report (Image Processing Institute, University of Southern California, Los Angeles, 1975)
A. Ron, Z. Shen, Affine systems in L2(Rd): the analysis of the analysis operator. J. Funct. Anal. 148(2), 408–447 (1997). doi:10.1007/BF02648888
C. Sutour, C. Deledalle, J. Aujol, Adaptive regularization of the NL-means: application to image and video denoising. IEEE Trans. Image Process. 23(8), 3506–3521 (2014). doi:10.1109/TIP.2014.2329448
X. Tai, C. Wu, Augmented Lagrangian method, dual methods and split Bregman iteration for ROF model, in Scale Space and Variational Methods in Computer Vision, Lecture Notes in Computer Science, vol. 5567 (Springer, Berlin, 2009), pp. 502–513. doi:10.1007/978-3-642-02256-2_42
C. Wu, J. Zhang, X. Tai, Augmented Lagrangian method for total variation restoration with non-quadratic fidelity. Inverse Probl. Imaging 5(1), 237–261 (2011). doi:10.3934/ipi.2011.5.237
Y. Xiao, T. Zeng, J. Yu, M. Ng, Restoration of images corrupted by mixed Gaussian-impulse noise via l1–l0 minimization. Pattern Recognit. 44(8), 1708–1720 (2011). doi:10.1016/j.patcog.2011.02.002
M. Yan, Restoration of images corrupted by impulse noise and mixed Gaussian impulse noise using blind inpainting. SIAM J. Imaging Sci. 6(3), 1227–1245 (2013). doi:10.1137/12087178X
J. Yang, Y. Zhang, W. Yin, An efficient TVL1 algorithm for deblurring multichannel images corrupted by impulsive noise. SIAM J. Sci. Comput. 31(4), 2842–2865 (2009). doi:10.1137/080732894
Z. Yang, M. Jacob, Nonlocal regularization of inverse problems: a unified variational framework. IEEE Trans. Image Process. 22(8), 3192–3203 (2013). doi:10.1109/TIP.2012.2216278
Acknowledgements
The author would like to thank the anonymous reviewers and the editor for their constructive comments and suggestions, which have improved greatly the development and presentation of this paper.
Author information
Authors and Affiliations
Corresponding author
Appendix: Convergence Analysis
Appendix: Convergence Analysis
In this section, we will give the global convergence analysis of the proposed ADMM algorithm for IR, i.e., Algorithm 1, in Theorem 1.
Before we show Theorem 1, we first change the constrained problem (15) into another equivalent form by firstly bringing in several notations. From here, we let \(X \triangleq {\mathbb {R}}^{M\times N}, Y \triangleq X \times X, Z \triangleq W(X) \subset X\), and \(A \triangleq [\nabla , W, K]^{\mathrm {T}}\) denotes an operator that maps X into \(Y \times Z \times X \), where \(\nabla \), W, and K map X into Y, Z, and X, respectively. After introducing two convex functionals \(F_(x)=0 : X \rightarrow {\mathbb {R}}\) and \(G_(\zeta ) \triangleq G([v, z, T]^{\mathrm {T}}) = \left\| T \right\| _1 + \alpha \sum \left\| v \right\| _2+\beta \left\| z \right\| _1 : Y \times Z \times X \rightarrow {\mathbb {R}}\), we give the equivalent formulation of problem (15) as follows
where \(\wp = [0, 0, y] \in Y \times Z \times X \). Then (17), the AL function of (15) turns into
where \(p = [p_1, p_2, p_3]^{\mathrm {T}} \in Y \times Z \times X \) and \(\lambda = [\lambda _1, \lambda _2, \lambda _3]^{\mathrm {T}} \in {\mathbb {R}} \times {\mathbb {R}} \times {\mathbb {R}}\). Note that \(\langle p, Ax-\zeta -\wp \rangle \) and \(\frac{\lambda }{2}\left\| Ax-\zeta -\wp \right\| _2^2\) mean doing operations per-component, i.e., \(\langle p_1, v-\nabla x \rangle + \langle p_2, z-Wx\rangle + \langle p_3, T-Kx+y \rangle \) and \(\frac{\lambda _1}{2}\left\| v-\nabla x \right\| _2^2\) \(+\frac{\lambda _2}{2}\left\| z-Wx \right\| _2^2 + \frac{\lambda _3}{2}\left\| T-Kx+y \right\| _2^2 \) and so do the other operations.
Now, we give the convergence theorem of Algorithm 1.
Theorem 1
Under the following two assumptions, the sequence \(\left\{ (x^k, \zeta ^k, p^k)\right\} \) generated by Algorithm 1 converges to a KKT point \((x^*, \zeta ^*, p^*)\) of (41).
Assumption 1
There exists a saddle point \(\theta ^* \triangleq (x^*,\zeta ^*,p^*)\) to problem (41), namely \(x^*\), \(\zeta ^*\), and \(p^*\) satisfy the KKT conditions
where F(x) and \(G(\zeta )\) are both convex functionals, hence, there exist two scalars \(\tau _F\) and \(\tau _G\) satisfy the following two inequalities, respectively.
Assumption 2
\(2-\xi >(1-\xi )^2.\)
Before giving the proof of Theorem 1, for simplicity, we straighten the image \(x \in {\mathbb {R}}^{M\times N}\) into a vector sill denoted as \(x \in {\mathbb {R}}^{MN}\) by stacking the columns of x and so do the other variables such as \(\zeta \) and p. Therefore, we have \(\zeta ,~p \in {\mathbb {R}}^{3MN}\), A is an operator that maps \({\mathbb {R}}^{MN}\) into \({\mathbb {R}}^{3MN}\) and \(A^{\mathrm {T}}A \in {\mathbb {R}}^{MN \times MN}\). Moreover, we let \(\left\| x \right\| _\Xi ^2 \triangleq x^{\mathrm {T}}\Xi x\), where \(\Xi \in {\mathbb {R}}^{MN \times MN}\). Furthermore, we introduce the following symbols
where \(\theta ^*\) is a KKT point, \(\theta ^k\) is the current point, \(\widehat{\theta }\) is the next point as if \(\xi =1\), and
where \({\mathbf {I}}\) and \(\mathbf {0}\) refer to the identity matrix and the zero matrix with proper size, respectively. From now on, we also denoted \(\lambda \triangleq \lambda _1=\lambda _2\) for simplicity.
From these definitions, it follows
and \(G \ge 0\); hence, G is a semi-norm.
Proof
We divide this process into the following three parts. \(\square \)
Step 1 To show the boundedness of the sequence \(\left\{ \theta ^k\right\} \).
The optimality conditions of the subproblems of Algorithm 1 are
By the convexity of G and the first optimality condition in Theorem 1, and (51), it follows that
Similarly, by the convexity of g, the second optimality condition in Theorem 1, and (52), we have
In addition, it follows the third optimality condition in Theorem 1 and (46) that
Then, adding (53) and (54) and using (55) give
which can be simplified as
By rearranging the terms, we have
From the relationship (50), we get the following identity
By (58), we have
and thus it follows
where
Then, \(\left\| \theta ^k-\theta ^* \right\| _G^2\) is bounded, as a result of which we have the boundedness of \({(x^k,p^k)}\) and that of \({\zeta ^k}\) by (55).
Step 2 To show that for a certain \(\rho >0\), \(\left\| \theta ^k-\theta ^* \right\| _G^2+\frac{\lambda }{\rho }\left\| \varUpsilon ^k \right\| _2^2\) is monotonically nonincreasing, where \(\varUpsilon ^k = Ax^k-\zeta ^k-\wp \) is the residual at iteration k, instead of which, we turn to proof that there exist \(\varpi >0\) such that
We first derive a lower bound for the cross-term \((p^k-{\widehat{p}})^{\mathrm {T}}A(x^k-x^{k+1})\). Combining (52) with the definition of \({\widehat{p}}\), i.e., (46), we have
The difference of the two terms on the left in (64) is
By (64), (65), and (46), we get
to which applying the Cauchy–Schwarz inequality gives
Substituting (67) into (61) and using the definition of \({\widehat{p}}\), we have
To prove such \(\varpi >0\) exists for (63), we only need the existence of \(\rho >0\) such that \(2-\xi -\frac{1}{\rho }\) and \(1-(1-\xi )^2\rho >0\), which holds if and only if \(2-\xi >(1-\xi )^2\), i.e., \(\xi \in (0,\frac{1+\sqrt{5}}{2})\).
Step 3 To show the global convergence of Algorithm 1.
Being bounded, \(\left\{ \theta ^k\right\} \) has a converging subsequence \(\left\{ \theta ^{k_j}\right\} \). Let \(\bar{\theta }=\lim _{j\rightharpoondown \infty }\theta ^{k_j}\). Next we will show \(\bar{\theta }\) is a KKT point. Let \(\theta ^*\) denote an arbitrary KKT point.
As we know that \(\left\| \theta ^k-\theta ^* \right\| _G^2+\frac{\lambda }{\rho }\left\| \varUpsilon ^k \right\| _2^2\) is monotonically nonincreasing and thus converging. Due to \(\varpi >0\) , \( \left\| \theta ^k-\theta ^{k+1} \right\| _G^2= \lambda \left\| A(x^k-x^{k+1}) \right\| _2^2+\frac{1}{\lambda \xi } \left\| p^k-p^{k+1} \right\| _2^2 \rightarrow 0\), so \(x^k-x^{k+1}\rightarrow 0\) and \(p^k-p^{k+1}\rightarrow 0\) and the following formula holds
Consequently, \( \left\| \theta ^k-\theta ^* \right\| _G^2\) also converges.
By passing limit on (69) over the subsequence, we have
Now on both sides of (51) and (52) taking limit over the subsequence, we obtain \(-\bar{p} \in \partial G(\bar{\zeta })\), \(A^{\mathrm {T}}\bar{p} \in \partial F(\bar{x})\). Then, together with (70), \(\bar{\theta }\) is a KKT point of problem (41) and now we can let \(\theta ^*=\bar{\theta }\). From \(\left\{ \theta ^{k_j}\right\} \rightarrow \bar{\theta }\) in j and the convergence of \(\parallel \theta ^k-\theta ^*\parallel _G^2\) it follows \(\left\| \theta ^k-\theta ^* \right\| _G^2\rightarrow 0 \) in k, which implies that \(p^k\rightarrow p^*\), \(x^k\rightarrow x^*\) and \(\zeta ^k\rightarrow \zeta ^*\) following from (69) and (70).
Then, we have finished the proof of Theorem 1. \(\square \)
Rights and permissions
About this article
Cite this article
Jiang, D. A Multi-parameter Regularization Model for Deblurring Images Corrupted by Impulsive Noise. Circuits Syst Signal Process 36, 3702–3730 (2017). https://doi.org/10.1007/s00034-016-0478-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00034-016-0478-1