Abstract
Denoising is an important preprocessing step that can improve the quality of the data and make it more suitable for further analysis, enhance the performance of machine learning models, identify underlying patterns, reduce computation time, and make data more interpretable by humans. Here we propose a tensor denoising approach based on Pareto efficient pairs and its relation with dual norms. We relate the problem of tensor denoising to that of maximizing the norm of the clean part while minimizing the norm of the noise. We propose a simple efficient method to remove additive noise of signals and compare the results, in terms of PSNR and MSE, with those of standard decomposition-based denoising methods over synthetically generated data.
This is a preview of subscription content, log in via an institution to check access.
Data availability
The author confirms that no real data have been used in this manuscript. All data are synthetically generated.
References
Derksen, H.: A general theory of singular values with applications to signal denoising. SIAM J. Appl. Algebra Geometry 2(4), 535–596 (2018)
Aurentz, J.L., Mach, T., Robol, L., Vandebril, R., Watkins, D.S.: Core-chasing algorithms for the eigenvalue problem. SIAM (2018)
Xia, D., Zhou, F.: The sup-norm perturbation of hosvd and low rank tensor denoising. J. Mach. Learn. Res. 20(1), 2206–2247 (2019)
Nie, Y., Chen, L., Zhu, H., Du, S., Yue, T., Cao, X.: Graph-regularized tensor robust principal component analysis for hyperspectral image denoising. Appl. Opt. 56(22), 6094–6102 (2017)
Hyvarinen, A., Karhunen, J., Oja, E.: Independent component analysis. Stud. Informat. Control 11(2), 205–207 (2002)
Li, H., Smith, S.M., Gruber, S., Lukas, S.E., Silveri, M.M., Hill, K.P., Killgore, W.D., Nickerson, L.D.: Denoising scanner effects from multimodal mri data using linked independent component analysis. Neuroimage 208, 116388 (2020)
Hosono, K., Ono, S., Miyata, T.: Weighted tensor nuclear norm minimization for color image denoising. In: 2016 IEEE International Conference on Image Processing (ICIP), pp. 3081–3085, IEEE (2016)
Zheng, Y.-B., Huang, T.-Z., Zhao, X.-L., Chen, Y., He, W.: Double-factor-regularized low-rank tensor factorization for mixed noise removal in hyperspectral image. IEEE Trans. Geosci. Remote Sens. 58(12), 8450–8464 (2020)
Huang, J., Cui, L.: Tensor singular spectrum decomposition: Multisensor denoising algorithm and application. IEEE Trans. Instrument. Measure. 72, 1–15 (2023)
Fan, L., Zhang, F., Fan, H., Zhang, C.: Brief review of image denoising techniques. Visual Comput. Ind. Biomed. Art 2(1), 1–12 (2019)
Kolda, T.G., Sun, J.: Scalable tensor decompositions for multi-aspect data mining. In: 2008 Eighth IEEE international conference on data mining, pp. 363–372, IEEE (2008)
Hillar, C.J., Lim, L.-H.: Most tensor problems are np-hard. J. ACM (JACM) 60(6), 1–39 (2013)
Kolda, T.G., Bader, B.W.: Tensor decompositions and applications. SIAM Rev. 51(3), 455–500 (2009)
Zhang, Z., Aeron, S.: Exact tensor completion using t-svd. IEEE Trans. Signal Process. 65(6), 1511–1526 (2016)
Kilmer, M.E., Braman, K., Hao, N., Hoover, R.C.: Third-order tensors as operators on matrices: A theoretical and computational framework with applications in imaging. SIAM J. Matrix Anal. Appl. 34(1), 148–172 (2013)
Markovsky, I.: Low Rank Approximation: Algorithms, Implementation, Applications, Vol. 906. Springer (2012)
Håstad, J.: Tensor rank is np-complete. In: International Colloquium on Automata, Languages, and Programming, pp. 451–460, Springer (1989)
Chinchuluun, A., Pardalos, P.M., Migdalas, A., Pitsoulis, L.: Pareto Optimality, Game Theory and Equilibria. Springer (2008)
Derksen, H.: On the nuclear norm and the singular value decomposition of tensors. Found. Comput. Math. 16(3), 779–811 (2016)
Chen, B., Li, Z.: On the tensor spectral p-norm and its dual norm via partitions. Comput. Opt. Appl. 75(3), 609–628 (2020)
F. Bach, “Convex relaxations of structured matrix factorizations,” arXiv preprint arXiv:1309.3117, 2013
Zhou, P., Lu, C., Lin, Z.: Tensor principal component analysis. In: Tensors for Data Processing, pp. 153–213, Elsevier (2022)
Friedland, S., Lim, L.-H.: Nuclear norm of higher-order tensors. Math. Comput. 87(311), 1255–1281 (2018)
S. Xue, W. Qiu, F. Liu, and X. Jin, “Low-rank tensor completion by truncated nuclear norm regularization,” in 2018 24th International Conference on Pattern Recognition (ICPR), pp. 2600–2605, IEEE, 2018
De Lathauwer, L., De Moor, B., Vandewalle, J.: On the best rank-1 and rank-(r 1, r 2,..., rn) approximation of higher-order tensors. SIAM J. Matrix Anal. Appl. 21(4), 1324–1342 (2000)
Acknowledgements
The author would like to express their sincere gratitude to the anonymous referees for their valuable feedback and constructive comments, which greatly improved the quality of this paper.
Author information
Authors and Affiliations
Corresponding author
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
Bagherian, M. Tensor denoising via dual Schatten norms. Optim Lett (2023). https://doi.org/10.1007/s11590-023-02068-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11590-023-02068-8