×

Spectral decompositions using one-homogeneous functionals. (English) Zbl 1361.35123

Summary: This paper discusses the use of absolutely one-homogeneous regularization functionals in a variational, scale space, and inverse scale space setting to define a nonlinear spectral decomposition of input data. We present several theoretical results that explain the relation between the different definitions. Additionally, results on the orthogonality of the decomposition, a Parseval-type identity, and the notion of generalized (nonlinear) eigenvectors closely link our nonlinear multiscale decompositions to the well-known linear filtering theory. Numerical results are used to illustrate our findings.

MSC:

35P30 Nonlinear eigenvalue problems and nonlinear spectral theory for PDEs
34L05 General spectral theory of ordinary differential operators
49R05 Variational methods for eigenvalues of operators
47A75 Eigenvalue problems for linear operators

References:

[1] F. Andreu, C. Ballester, V. Caselles, and J. M. Mazón, {\it Minimizing total variation flow}, Differential Integral Equations, 14 (2001), pp. 321-360. · Zbl 1020.35037
[2] F. Andreu, V. Caselles, J. I. D\iaz, and J. M. Mazón, {\it Some qualitative properties for the total variation flow}, J. Functional Anal., 188 (2002), pp. 516-547. · Zbl 1042.35018
[3] G. Aubert and P. Kornprobst, {\it Mathematical Problems in Image Processing}, Appl. Math. Sci. 147, Springer-Verlag, Berlin, 2002. · Zbl 1109.35002
[4] S. Bartels, R. H. Nochetto, J. Abner, and A. J. Salgado, {\it Discrete Total Variation Flows Without Regularization}, arXiv:1212.1137, 2012. · Zbl 1292.65106
[5] G. Bellettini, V. Caselles, and M. Novaga, {\it The total variation flow in \(R^N\)}, J. Differential Equations, 184 (2002), pp. 475-525. · Zbl 1036.35099
[6] M. Benning, C. Brune, M. Burger, and J. Müller, {\it Higher-order tv methods: Enhancement via Bregman iteration}, J. Sci. Comput., 54 (2013), pp. 269-310. · Zbl 1308.94012
[7] M. Benning and M. Burger, {\it Ground states and singular vectors of convex variational regularization methods}, Methods. Appl. Anal., 20 (2013), pp. 295-334. · Zbl 1294.49020
[8] K. Bredies, K. Kunisch, and T. Pock, {\it Total generalized variation}, SIAM J. Imaging Scie., 3 (2010), pp. 492-526. · Zbl 1195.49025
[9] T. Brox, M. Welk, G. Steidl, and J. Weickert, {\it Proceedings of the 4th International Conference on Scale Space Methods in Computer Vision}, Springer, Berlin, 2003, pp. 86-100. · Zbl 1067.68722
[10] M. Burger, L. Eckardt, G. Gilboa, and M. Moeller, Spectral representations of one-homogeneous functionals, in {\it Proceedings of the 5th International Conference on Scale Space and Variational Methods in Computer Vision,} Springer, Berlin, 2015, pp. 16-27. · Zbl 1361.35123
[11] M. Burger, K. Frick, S. Osher, and O. Scherzer, {\it Inverse total variation flow}, Multiscale Model. Simul., 6 (2007), pp. 366-395. · Zbl 1147.49026
[12] M. Burger, G. Gilboa, S. Osher, and J. Xu, {\it Nonlinear inverse scale space methods}, Commun. Math. Sci., 4 (2006), pp. 179-212. · Zbl 1106.68117
[13] A. Chambolle and P. L. Lions, {\it Image recovery via total variation minimization and related problems}, Numeri Mathe., 76 (1997), pp. 167-188. · Zbl 0874.68299
[14] L. Dorst and R. Van den Boomgaard, {\it Morphological signal processing and the slope transform}, Signal Process., 38 (1994), pp. 79-98.
[15] N. Dunford and J. T. Schwartz, {\it Linear Operators. Part 2: Spectral Theory. Self Adjoint Operators in Hilbert Space}, Interscience Publishers, New York, 1963. · Zbl 0128.34803
[16] J. Duran, M. Moeller, C. Sbert, and D. Cremers, {\it Collaborative total variation: A general framework for vectorial TV models}, SIAM J. Imaging Sci., 9 (2016), pp. 116-151. · Zbl 1381.94016
[17] V. Duval and G. Peyré, {\it Exact support recovery for sparse spikes deconvolution}, Found. Comput. Math., 15 (2015), pp. 1315-1355. · Zbl 1327.65104
[18] L. C. Evans, {\it Partial Differential Equations}, Grad. Stud, Math. 19, AMS, Prvidence, RI, 1991.
[19] J.-J. Fuchs, {\it Spread Representations}, in Proceedings of the Asilomar Conference on Signals, Systems, and Computers, 2011.
[20] Y. Giga and R. V. Kohn, {\it Scale-invariant Extinction Time Estimates for Some Singular Diffusion Equations}, Preprint Series in Mathematics, Hokkaido University, 2010. · Zbl 1223.35211
[21] G. Gilboa, {\it A Spectral Approach to Total Variation}, A. Kuijper et al., eds., Lecture Notes in Comput. Sci. 7893, Springer, New York, 2013, pp. 36-47. · Zbl 1484.94007
[22] G. Gilboa, {\it A total variation spectral framework for scale and texture analysis}, SIAM J. Imaging Sci., 7 (2014), pp. 1937-1961. · Zbl 1361.94014
[23] G. Gilboa, M. Moeller, and M. Burger, {\it Nonlinear spectral analysis via one-homogeneous functionals–overview and future prospects}, J. Math. Imaging Vision, 56 (2016), pp. 300-319. · Zbl 1409.94182
[24] B. Grünbaum, {\it Convex Polytopes}, Grad. Texts Math., Springer, Berlin, 2003. · Zbl 1024.52001
[25] D. Horesh, {\it Separation Surfaces in the Spectral TV Domain for Texture Decomposition}, Masters thesis, Technion, Haifa, Israel, 2015. · Zbl 1408.94251
[26] D. Horesh and G. Gilboa, {\it Separation surfaces in the spectral TV domain for texture decomposition}, IEEE Trans. Image Process., 25 (2016), pp. 4260-4270. · Zbl 1408.94251
[27] U. Köthe, {\it Local Appropriate Scale in Morphological Scale-Space}, in ECCV’96, Springer, Berlin, 1996, pp. 219-228.
[28] J. Mairal, M. Elad, and G. Sapiro, {\it Sparse representation for color image restoration}, IEEE Trans. Image Process, 17 (2008), pp. 53-69. · Zbl 1194.49041
[29] Y. Meyer, {\it Oscillating Patterns in Image Processing and Nonlinear Evolution Equations: The Fifteenth Dean Jacqueline B. Lewis Memorial Lectures}, AMS, Providence, RI, Boston, MA, 2001. · Zbl 0987.35003
[30] M. Moeller and M. Burger, {\it Multiscale methods for polyhedral regularizations}, SIAM J. Optim., 23 (2013), pp. 1424-1456. · Zbl 1281.65091
[31] M. Moeller, J. Diebold, G. Gilboa, and D. Cremers, {\it Learning Nonlinear Spectral Filters for Color Image Reconstruction}, in Proceedings of ICCV, 2015.
[32] J. Müller, {\it Advanced Image Reconstruction and Denoising: Bregmanized (Higher Order) Total Variation and Application in PET}, Ph.D. thesis, University of Münster, 2013. · Zbl 1271.94001
[33] M. R. Osborne, B. Presnell, and B. A. Turlach, {\it A new approach to variable selection in least squares problems}, IMA J. Numer. Anal., 20 (2000), pp. 389-403. · Zbl 0962.65036
[34] S. Osher, M. Burger, D. Goldfarb, J. Xu, and W. Yin, {\it An iterative regularization method for total variation based image restoration}, SIAM J. Multiscale Model. Simul., 4 (2005), pp. 460-489. · Zbl 1090.94003
[35] L. Rudin, S. Osher, and E. Fatemi, {\it Nonlinear total variation based noise removal algorithms}, Phys. D, 60 (1992), pp. 259-268. · Zbl 0780.49028
[36] J. Sporring and J. Weickert, {\it Bounds on shape recognition performance}, IEEE Trans. Image Process. 45 (1999), pp. 1151-1158.
[37] G. Steidl, J. Weickert, T. Brox, P. Mrázek, and M. Welk, {\it On the equivalence of soft wavelet shrinkage, total variation diffusion, total variation regularization, and SIDEs}, SIAM J. Numer. Anal., 42 (2004), pp. 686-713. · Zbl 1083.94001
[38] R. J. Tibshirani and J. Taylor, {\it The solution path of the generalized lasso}, Ann. Statist., 39 (2011), pp. 1335-1371. · Zbl 1234.62107
[39] S. Vaiter, G. Peyré, and J. M. Fadili, {\it Model Consistency of Partly Smooth Regularizers}, arXiv:1405.1004, 2014. · Zbl 1464.62345
[40] J. Weickert, {\it Anisotropic Diffusion in Image Processing}, Teubner-Verlag, Stuttgart, Germany, 1998. · Zbl 0886.68131
[41] M. Welk, G. Steidl, and J. Weickert, {\it Locally analytic schemes: A link between diffusion filtering and wavelet shrinkage}, Appl. Comput. Harmon. Anal., 24 (2008), pp. 195-224. · Zbl 1161.68831
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.