Abstract
Hyperspectral image processing methods based on Tucker decomposition by utilizing low-rank and sparse priors are sensitive to the model order, and merely utilizing the global structural information. After statistical analysis on hyperspectral images, we find that the smoothness underlying hyperspectral image encoding local structural information is ubiquity in each mode. Based on this observation, we propose a novel smooth coupled Tucker decomposition scheme with two smoothness constraints imposed on the subspace factor matrices to reveal the local structural information of hyperspectral image. In addition, efficient algorithms are designed and experimental results demonstrate the effectiveness of selecting optimal model order for hyperspectral image super-resolution due to the integration of the subspace smoothness.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In this study, the rank of CPD/Tucker decomposition and the size of dictionaries in [4] are collectively referred as model order. In our models, the size of core tensor is model order.
1. References
Bu, Y., et al.: Hyperspectral and multispectral image fusion via graph laplacian-guided coupled tensor decomposition. IEEE Trans. Geosci. Remote Sens. 59(1), 648–662 (2020)
Xue, J., Zhao, Y.Q., Bu, Y., Liao, W., Chan, J.C.W., Philips, W.: Spatial-spectral structured sparse low-rank representation for hyperspectral image super-resolution. IEEE Trans. Image Process. 30, 3084–3097 (2021)
Li, H., Li, W., Han, G., Liu, F.: Coupled tensor decomposition for hyperspectral pansharpening. IEEE Access 6, 34 206–34 213 (2018)
Li, S., Dian, R., Fang, L., Bioucas-Dias, J.M.: Fusing hyperspectral and multispectral images via coupled sparse tensor factorization. IEEE Trans. Image Process. 27(8), 4118–4130 (2018)
Kanatsoulis, C.I., Fu, X., Sidiropoulos, N.D., Ma, W.-K.: Hyperspectral super-resolution: a coupled tensor factorization approach. IEEE Trans. Signal Process. 66(24), 6503–6517 (2018)
Prevost, C., Usevich, K., Comon, P., Brie, D.: Hyperspectral super-resolution with coupled tucker approximation: Identifiability and SVD-based algorithms, November 2018, working paper or preprint. https://hal.archives-ouvertes.fr/hal-01911969
Prvost, C., Usevich, K., Comon, P., Brie, D.: Coupled tensor low-rank multilinear approximation for hyperspectral super-resolution. In: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 5536–5540 (2019)
Wang, B., Yang, L., Zhao, Y.: Polo: learning explicit cross-modality fusion for temporal action localization. IEEE Signal Process. Lett. 28, 503–507 (2021)
Wald, L., Ranchin, T., Mangolini, M.: Fusion of satellite images of different spatial resolutions: assessing the quality of resulting images. Photogramm. Eng. Rem. S. 63(6), 691–699 (1997)
Yokoya, N., Grohnfeldt, C., Chanussot, J.: Hyperspectral and multispectral data fusion: a comparative review of the recent literature. IEEE Geosci. Remote Sens. Mag. 5(2), 29–56 (2017)
Xue, J., Zhao, Y., Liao, W., Chan, J.C., Kong, S.G.: Enhanced sparsity prior model for low-rank tensor completion. IEEE Trans. Neural Netw. Learn. Syst. 1–15 (2019)
Salakhutdinov, R., Mnih, A.: Probabilistic matrix factorization. In: Proceedings Advanced International Conference Neural Information Processing System, pp. 1257–1264 (2007)
Yokota, T., Zhao, Q., Cichocki, A.: Smooth PARAFAC decomposition for tensor completion. IEEE Trans. Signal Process. 64(20), 5423–5436 (2016)
Imaizumi, M., Hayashi, K.: Tensor decomposition with smoothness, ser. Proceedings of Machine Learning Research. Precup, D., Teh, Y.W. (eds.) vol. 70. International Convention Centre, Sydney, Australia: PMLR, 06–11 Aug 2017, pp. 1597–1606 (2017). http://proceedings.mlr.press/v70/imaizumi17a.html
Attouch, H., Bolte, J., Redont, P., Soubeyran, A.: Proximal alternating minimization and projection methods for nonconvex problems: an approach based on the kurdyka-łojasiewicz inequality. Math. Oper. Res. 35(2), 438–457 (2010)
Huang, K., Sidiropoulos, N.D., Liavas, A.P.: A flexible and efficient algorithmic framework for constrained matrix and tensor factorization. IEEE Trans. Signal Process. 64(19), 5052–5065 (2016)
Boyd, S., Parikh, N., Chu, E., Peleato, B., Eckstein, J., et al.: Distributed optimization and statistical learning via the alternating direction method of multipliers. Foundat. Trends Mach. Learn. 3(1), 1–122 (2011)
Dell’Acqua, F., Gamba, P., Ferrari, A., Palmason, J.A., Benediktsson, J.A., Arnason, K.: Exploiting spectral and spatial information in hyperspectral urban data with high resolution. IEEE Geosci. Remote Sens. Lett. 1(4), 322–326 (2004)
Wei, Q., Dobigeon, N., Tourneret, J.-Y.: Fast fusion of multi-band images based on solving a sylvester equation. IEEE Trans. Image Process. 24(11), 4109–4121 (2015)
Acknowledgements
This work was supported in part by the National Natural Science Foundation of China under Grant 61371152 and Grant 61771391, in part by the Shenzhen Municipal Science and Technology Innovation Committee under Grant JCYJ20170815162956949 and JCYJ20180306171146740 and in part by Key R&D Plan of Shaanxi Province 2020ZDLGY07–11, by Natural Science Basic Research plan in Shaanxi Province of China 2018JM6056 and by the innovation Foundation for Doctor Dissertation of Northwestern Polytechnical University CX2021081.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Bu, Y., Zhao, Y., Xue, J., Chan, J.CW. (2021). Smooth Coupled Tucker Decomposition for Hyperspectral Image Super-Resolution. In: Ma, H., et al. Pattern Recognition and Computer Vision. PRCV 2021. Lecture Notes in Computer Science(), vol 13021. Springer, Cham. https://doi.org/10.1007/978-3-030-88010-1_20
Download citation
DOI: https://doi.org/10.1007/978-3-030-88010-1_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-88009-5
Online ISBN: 978-3-030-88010-1
eBook Packages: Computer ScienceComputer Science (R0)