Figure 1.
Decomposition of the original image (left) in structural component $ u $ (middle) and textural component $ v $ (right)
-
Previous Article
Commutative properties for conservative space-time DG discretizations of optimal control problems involving the viscous Burgers equation
- MCRF Home
- This Issue
-
Next Article
Local null controllability of the penalized Boussinesq system with a reduced number of controls
doi: 10.3934/mcrf.2021048
Online First
Online First articles are published articles within a journal that have not yet been assigned to a formal issue. This means they do not yet have a volume number, issue number, or page numbers assigned to them, however, they can still be found and cited using their DOI (Digital Object Identifier). Online First publication benefits the research community by making new scientific discoveries known as quickly as possible.
Readers can access Online First articles via the “Online First” tab for the selected journal.
Parameter learning and fractional differential operators: Applications in regularized image denoising and decomposition problems
Department of Applied Mathematics, Mathematical Institute, University of Freiburg, Hermann-Herder-Str. 10, 79104 Freiburg i. Br., Germany |
In this paper, we focus on learning optimal parameters for PDE-based image denoising and decomposition models. First, we learn the regularization parameter and the differential operator for gray-scale image denoising using the fractional Laplacian in combination with a bilevel optimization problem. In our setting the fractional Laplacian allows the use of Fourier transform, which enables the optimization of the denoising operator. We prove stable and explainable results as an advantage in comparison to machine learning approaches. The numerical experiments correlate with our theoretical model settings and show a reduction of computing time in contrast to the Rudin-Osher-Fatemi model. Second, we introduce a new regularized image decomposition model with the fractional Laplacian and the Riesz potential. We provide an explicit formula for the unique solution and the numerical experiments illustrate the efficiency.
Keywords:
Fractional Laplacian,
bilevel optimization,
parameter learning,
partial differential equations,
variational regularization.
Mathematics Subject Classification:
Primary: 49J20, 49K20, 49N45; Secondary: 65N12, 65N35.
Citation:
Sören Bartels, Nico Weber.
Parameter learning and fractional differential operators: Applications in regularized image denoising and decomposition problems. Mathematical Control & Related Fields, doi: 10.3934/mcrf.2021048
![]()
References:
| [1] |
H. Antil and S. Bartels,
Spectral approximation of fractional PDEs in image processing and phase field modeling, Comput. Methods Appl. Math., 17 (2017), 661-678.
doi: 10.1515/cmam-2017-0039. |
| [2] |
H. Antil, Z. W. Di and R. Khatri, Bilevel optimization, deep learning and fractional laplacian regularization with applications in tomography, Inverse Problems, 36 (2020), 064001.
doi: 10.1088/1361-6420/ab80d7. |
| [3] |
H. Antil, E. Otárola and A. J. Salgado,
Optimization with respect to order in a fractional diffusion model: Analysis, approximation and algorithmic aspects, J. Sci. Comput., 77 (2018), 204-224.
doi: 10.1007/s10915-018-0703-0. |
| [4] |
H. Antil and C. N. Rautenberg,
Sobolev spaces with non-Muckenhoupt weights, fractional elliptic operators, and applications, SIAM J. Math. Anal., 51 (2019), 2479-2503.
doi: 10.1137/18M1224970. |
| [5] |
J.-F. Aujol, G. Gilboa, T. Chan and S. Osher,
Structure-texture image decomposition——modeling, algorithms, and parameter selection, International Journal of Computer Vision, 67 (2006), 111-136.
doi: 10.1007/s11263-006-4331-z. |
| [6] |
J. Batson and L. Royer, {N}oise2{S}elf: Blind denoising by self-supervision, in Proceedings of the 36th International Conference on Machine Learning (eds. K. Chaudhuri and R. Salakhutdinov), vol. 97 of Proceedings of Machine Learning Research, PMLR, (2019), 524–533. Google Scholar |
| [7] |
S. Boyd and L. Vandenberghe, Convex Optimization, Cambridge University Press, Cambridge, 2004.
doi: 10.1017/CBO9780511804441.
Google Scholar
|
| [8] |
F. Boyer and P. Fabrie, Mathematical Tools for the Study of the Incompressible Navier-Stokes Equations and Related Models, vol. 183 of Applied Mathematical Sciences, Springer, New York, 2013.
doi: 10.1007/978-1-4614-5975-0. |
| [9] |
A. Bueno-Orovio, D. Kay and K. Burrage,
Fourier spectral methods for fractional-in-space reaction-diffusion equations, BIT, 54 (2014), 937-954.
doi: 10.1007/s10543-014-0484-2. |
| [10] |
Y. Gousseau and J.-M. Morel,
Are natural images of bounded variation?, SIAM J. Math. Anal., 33 (2001), 634-648.
doi: 10.1137/S0036141000371150. |
| [11] |
K. Kunish and T. Pock,
A bilevel optimization approach for parameter learning in variational models, SIAM Journal on Imaging Sciences, 6 (2013), 938-983.
doi: 10.1137/120882706. |
| [12] |
P. Liu and C.-B. Sch{ö}nlieb, Learning optimal orders of the underlying euclidean norm in total variation image denoising, arXiv preprint arXiv: 1903.11953. Google Scholar |
| [13] |
Q. Liu, Z. Zhang and Z. Guo,
On a fractional reaction-diffusion system applied to image decomposition and restoration, Comput. Math. Appl., 78 (2019), 1739-1751.
doi: 10.1016/j.camwa.2019.05.030. |
| [14] |
Y. Nesterov, Introductory Lectures on Convex Optimization, Springer US, 2004.
doi: 10.1007/978-1-4419-8853-9. |
| [15] |
S. Osher, A. Solé and L. Vese,
Image decomposition and restoration using total variation minimization and the $H^{-1}$ norm, Multiscale Model. Simul., 1 (2003), 349-370.
doi: 10.1137/S1540345902416247. |
| [16] |
G. Peyré,
The numerical tours of signal processing, Computing in Science & Engineering, 13 (2011), 94-97.
doi: 10.1109/MCSE.2011.71. |
| [17] |
L. I. Rudin, S. Osher and E. Fatemi,
Nonlinear total variation based noise removal algorithms, Phys. D, 60 (1992), 259-268.
doi: 10.1016/0167-2789(92)90242-F. |
| [18] |
J. Saranen and G. Vainikko, Periodic Integral and Pseudodifferential Equations with Numerical Approximation, Springer Monographs in Mathematics, Springer-Verlag, Berlin, 2002.
doi: 10.1007/978-3-662-04796-5. |
| [19] |
J. Sprekels and E. Valdinoci,
A new type of identification problems: Optimizing the fractional order in a nonlocal evolution equation, SIAM J. Control Optim., 55 (2017), 70-93.
doi: 10.1137/16M105575X. |
| [20] |
D. Ulyanov, A. Vedaldi and V. Lempitsky,
Deep image prior, International Journal of Computer Vision volume, 128 (2020), 1867-1888.
doi: 10.1007/s11263-020-01303-4. |
show all references
References:
| [1] |
H. Antil and S. Bartels,
Spectral approximation of fractional PDEs in image processing and phase field modeling, Comput. Methods Appl. Math., 17 (2017), 661-678.
doi: 10.1515/cmam-2017-0039. |
| [2] |
H. Antil, Z. W. Di and R. Khatri, Bilevel optimization, deep learning and fractional laplacian regularization with applications in tomography, Inverse Problems, 36 (2020), 064001.
doi: 10.1088/1361-6420/ab80d7. |
| [3] |
H. Antil, E. Otárola and A. J. Salgado,
Optimization with respect to order in a fractional diffusion model: Analysis, approximation and algorithmic aspects, J. Sci. Comput., 77 (2018), 204-224.
doi: 10.1007/s10915-018-0703-0. |
| [4] |
H. Antil and C. N. Rautenberg,
Sobolev spaces with non-Muckenhoupt weights, fractional elliptic operators, and applications, SIAM J. Math. Anal., 51 (2019), 2479-2503.
doi: 10.1137/18M1224970. |
| [5] |
J.-F. Aujol, G. Gilboa, T. Chan and S. Osher,
Structure-texture image decomposition——modeling, algorithms, and parameter selection, International Journal of Computer Vision, 67 (2006), 111-136.
doi: 10.1007/s11263-006-4331-z. |
| [6] |
J. Batson and L. Royer, {N}oise2{S}elf: Blind denoising by self-supervision, in Proceedings of the 36th International Conference on Machine Learning (eds. K. Chaudhuri and R. Salakhutdinov), vol. 97 of Proceedings of Machine Learning Research, PMLR, (2019), 524–533. Google Scholar |
| [7] |
S. Boyd and L. Vandenberghe, Convex Optimization, Cambridge University Press, Cambridge, 2004.
doi: 10.1017/CBO9780511804441.
Google Scholar
|
| [8] |
F. Boyer and P. Fabrie, Mathematical Tools for the Study of the Incompressible Navier-Stokes Equations and Related Models, vol. 183 of Applied Mathematical Sciences, Springer, New York, 2013.
doi: 10.1007/978-1-4614-5975-0. |
| [9] |
A. Bueno-Orovio, D. Kay and K. Burrage,
Fourier spectral methods for fractional-in-space reaction-diffusion equations, BIT, 54 (2014), 937-954.
doi: 10.1007/s10543-014-0484-2. |
| [10] |
Y. Gousseau and J.-M. Morel,
Are natural images of bounded variation?, SIAM J. Math. Anal., 33 (2001), 634-648.
doi: 10.1137/S0036141000371150. |
| [11] |
K. Kunish and T. Pock,
A bilevel optimization approach for parameter learning in variational models, SIAM Journal on Imaging Sciences, 6 (2013), 938-983.
doi: 10.1137/120882706. |
| [12] |
P. Liu and C.-B. Sch{ö}nlieb, Learning optimal orders of the underlying euclidean norm in total variation image denoising, arXiv preprint arXiv: 1903.11953. Google Scholar |
| [13] |
Q. Liu, Z. Zhang and Z. Guo,
On a fractional reaction-diffusion system applied to image decomposition and restoration, Comput. Math. Appl., 78 (2019), 1739-1751.
doi: 10.1016/j.camwa.2019.05.030. |
| [14] |
Y. Nesterov, Introductory Lectures on Convex Optimization, Springer US, 2004.
doi: 10.1007/978-1-4419-8853-9. |
| [15] |
S. Osher, A. Solé and L. Vese,
Image decomposition and restoration using total variation minimization and the $H^{-1}$ norm, Multiscale Model. Simul., 1 (2003), 349-370.
doi: 10.1137/S1540345902416247. |
| [16] |
G. Peyré,
The numerical tours of signal processing, Computing in Science & Engineering, 13 (2011), 94-97.
doi: 10.1109/MCSE.2011.71. |
| [17] |
L. I. Rudin, S. Osher and E. Fatemi,
Nonlinear total variation based noise removal algorithms, Phys. D, 60 (1992), 259-268.
doi: 10.1016/0167-2789(92)90242-F. |
| [18] |
J. Saranen and G. Vainikko, Periodic Integral and Pseudodifferential Equations with Numerical Approximation, Springer Monographs in Mathematics, Springer-Verlag, Berlin, 2002.
doi: 10.1007/978-3-662-04796-5. |
| [19] |
J. Sprekels and E. Valdinoci,
A new type of identification problems: Optimizing the fractional order in a nonlocal evolution equation, SIAM J. Control Optim., 55 (2017), 70-93.
doi: 10.1137/16M105575X. |
| [20] |
D. Ulyanov, A. Vedaldi and V. Lempitsky,
Deep image prior, International Journal of Computer Vision volume, 128 (2020), 1867-1888.
doi: 10.1007/s11263-020-01303-4. |
Figure 2.
Fractional Laplacian (1) of image $ u_d $ . A higher exponent $ s $ in (1) results in stronger smoothing of high oscillations
Figure 3.
Riesz potential (2) of image $ u_d $ . A lower exponent $ s $ in (2) smooths the image $ u_d $
Figure 4.
The denoising via solving (3) strongly depends on the choice of the parameters $ s $ and $ \alpha $
Figure 5.
Energy functional $ j_n $ with function $ \varphi_1 $ and different choices for $ s $ and $ \alpha $
Figure 6.
Results for the image "boat". As clearly seen, the model eliminates the fog based on the normally distributed noise
Figure 7.
In the image "peppers" there is a significant reduction of noise while maintaining the edges
1]. The small deviation of the optimal parameters is due to the influence of the function $ \varphi_1(s,\alpha) $ ">Figure 8.
With a higher resolution($ n = 1566 $ ) than in the sample images before ($ n = 512 $ ), the noise can be better suppressed optically. The denoised image is visually almost identical to the image in [1]. The small deviation of the optimal parameters is due to the influence of the function $ \varphi_1(s,\alpha) $
Figure 10.
We study the influence of the noise on the parameter $ s $ . The numerical results coincide with theoretical considerations that more noise implies a stronger smoothing of the image, i.e a higher value of the parameter $ s $
Figure 11.
In the case of the parameter $ \alpha $ the numerical results coincide with theoretical considerations. A higher noise has the result that the denoised image is more far away from the noisy image, i.e a lower value of the parameter $ \alpha $
Figure 12.
In comparison to the fractional Laplacian model, the ROF model produces smoother edges. Moreover the ROF model has a better SSIM-value
Figure 13.
Runtime comparison between the fractional Laplacian model and the ROF model. Both models show empirically a linear runtime, but the fractional Laplacian model has a reduced computing time by factors 16-22. We used MATLAB R2015a with CPU i3-3240 and 8 GB RAM
Figure 14.
The PSNR(peak signal-to-noise ratio) of the ROF model outperforms the Laplacian model, but the discrepancy depends on the choice of the image. We observe a decreasing PSNR for a larger variance, implying a stronger noise. A higher PSNR value implies a better image quality
Figure 15.
The SSIM(structural similarity index measure) of the ROF model outperforms the Laplacian model, again the performance gap highly depends on the image. We observe a decreasing SSIM for stronger noise. A higher SSIM value implies a better image quality
Figure 16.
Detail of the image "Baboon". The results of the Fractional Laplacian and ROF model are similar.
Figure 17.
Image components with fixed $ s_1 = 0.2, s_2 = -1, \beta = 1 $ and different choices for parameter $ \alpha $ for the noise-free image. The choice of the parameter $ \alpha $ has a strong impact on the decomposition
$ \bar{s_1} = 0.172,\; \bar{\alpha} = 9999.7,\; \bar{s_2} = -0.926 $ and $ \bar{\beta} = 10410 $ . We see a significant improvement of the SSIM-value and the desired image decomposition">Figure 18.
Decomposition of the image "kentaur" with optimal parameters $ \bar{s_1} = 0.172,\; \bar{\alpha} = 9999.7,\; \bar{s_2} = -0.926 $ and $ \bar{\beta} = 10410 $ . We see a significant improvement of the SSIM-value and the desired image decomposition
Figure 19.
The component $ u $ of (26) is obtained with optimal parameters $ \bar{s_1} = 0.351,\; \bar{\alpha} = 9999.4,\; \bar{s_2} = -0.918 $ and $ \bar{\beta} = 9997.2 $ . The fractional image decomposition model achieves a significantly better result than the fractional image denoising model with an improvement of the SSIM value by 0.4. The result is less blurred and has a better contrast
$ \sigma = 0.1 $). The model (29) can not sufficiently distinguish between noise and component $ v $ ">Figure 20.
Detail of the image "pepper" ($ \sigma = 0.1 $ ). The model (29) can not sufficiently distinguish between noise and component $ v $
$ \sigma = 0.1 $)">Figure 21.
Detail of the image "Baboon" ($ \sigma = 0.1 $ )
| [1] |
Tomasz Kosmala, Markus Riedle. Variational solutions of stochastic partial differential equations with cylindrical Lévy noise. Discrete & Continuous Dynamical Systems - B, 2021, 26 (6) : 2879-2898. doi: 10.3934/dcdsb.2020209 |
| [2] |
Youssef El Hadfi, Zhaosheng Feng, Abdelghani Ghazdali, Amine Laghrib. Preface to the special issue "Partial Differential Equations, Optimization and their Applications". Discrete & Continuous Dynamical Systems - S, 2022, 15 (1) : i-ii. doi: 10.3934/dcdss.2021163 |
| [3] |
Changpin Li, Zhiqiang Li. Asymptotic behaviors of solution to partial differential equation with Caputo–Hadamard derivative and fractional Laplacian: Hyperbolic case. Discrete & Continuous Dynamical Systems - S, 2021, 14 (10) : 3659-3683. doi: 10.3934/dcdss.2021023 |
| [4] |
Frank Pörner, Daniel Wachsmuth. Tikhonov regularization of optimal control problems governed by semi-linear partial differential equations. Mathematical Control & Related Fields, 2018, 8 (1) : 315-335. doi: 10.3934/mcrf.2018013 |
| [5] |
Liping Luo, Zhenguo Luo, Yunhui Zeng. New results for oscillation of fractional partial differential equations with damping term. Discrete & Continuous Dynamical Systems - S, 2021, 14 (9) : 3223-3231. doi: 10.3934/dcdss.2020336 |
| [6] |
Lorenzo Brasco, Enea Parini, Marco Squassina. Stability of variational eigenvalues for the fractional $p-$Laplacian. Discrete & Continuous Dynamical Systems, 2016, 36 (4) : 1813-1845. doi: 10.3934/dcds.2016.36.1813 |
| [7] |
Gernot Holler, Karl Kunisch. Learning nonlocal regularization operators. Mathematical Control & Related Fields, 2022, 12 (1) : 81-114. doi: 10.3934/mcrf.2021003 |
| [8] |
María J. Garrido–Atienza, Kening Lu, Björn Schmalfuss. Random dynamical systems for stochastic partial differential equations driven by a fractional Brownian motion. Discrete & Continuous Dynamical Systems - B, 2010, 14 (2) : 473-493. doi: 10.3934/dcdsb.2010.14.473 |
| [9] |
Bin Pei, Yong Xu, Yuzhen Bai. Convergence of p-th mean in an averaging principle for stochastic partial differential equations driven by fractional Brownian motion. Discrete & Continuous Dynamical Systems - B, 2020, 25 (3) : 1141-1158. doi: 10.3934/dcdsb.2019213 |
| [10] |
Dariusz Idczak. A Gronwall lemma for functions of two variables and its application to partial differential equations of fractional order. Mathematical Control & Related Fields, 2022, 12 (1) : 225-243. doi: 10.3934/mcrf.2021019 |
| [11] |
Omar Saber Qasim, Ahmed Entesar, Waleed Al-Hayani. Solving nonlinear differential equations using hybrid method between Lyapunov's artificial small parameter and continuous particle swarm optimization. Numerical Algebra, Control & Optimization, 2021, 11 (4) : 633-644. doi: 10.3934/naco.2021001 |
| [12] |
Farid Tari. Two-parameter families of implicit differential equations. Discrete & Continuous Dynamical Systems, 2005, 13 (1) : 139-162. doi: 10.3934/dcds.2005.13.139 |
| [13] |
Farid Tari. Two parameter families of binary differential equations. Discrete & Continuous Dynamical Systems, 2008, 22 (3) : 759-789. doi: 10.3934/dcds.2008.22.759 |
| [14] |
Pingping Niu, Shuai Lu, Jin Cheng. On periodic parameter identification in stochastic differential equations. Inverse Problems & Imaging, 2019, 13 (3) : 513-543. doi: 10.3934/ipi.2019025 |
| [15] |
Huan Han. A variational model with fractional-order regularization term arising in registration of diffusion tensor image. Inverse Problems & Imaging, 2018, 12 (6) : 1263-1291. doi: 10.3934/ipi.2018053 |
| [16] |
Herbert Koch. Partial differential equations with non-Euclidean geometries. Discrete & Continuous Dynamical Systems - S, 2008, 1 (3) : 481-504. doi: 10.3934/dcdss.2008.1.481 |
| [17] |
Lorenzo Zambotti. A brief and personal history of stochastic partial differential equations. Discrete & Continuous Dynamical Systems, 2021, 41 (1) : 471-487. doi: 10.3934/dcds.2020264 |
| [18] |
Wilhelm Schlag. Spectral theory and nonlinear partial differential equations: A survey. Discrete & Continuous Dynamical Systems, 2006, 15 (3) : 703-723. doi: 10.3934/dcds.2006.15.703 |
| [19] |
Eugenia N. Petropoulou, Panayiotis D. Siafarikas. Polynomial solutions of linear partial differential equations. Communications on Pure & Applied Analysis, 2009, 8 (3) : 1053-1065. doi: 10.3934/cpaa.2009.8.1053 |
| [20] |
Arnulf Jentzen. Taylor expansions of solutions of stochastic partial differential equations. Discrete & Continuous Dynamical Systems - B, 2010, 14 (2) : 515-557. doi: 10.3934/dcdsb.2010.14.515 |
2020 Impact Factor: 1.284
Tools
Article outline
Figures and Tables
[Back to Top]