Abstract
Hyperspectral images (HSIs) usually have high spectral resolution and low spatial resolution. 3D convolutional neural networks (3D-CNNs) can successfully extract joint spectral and spatial information. But these networks do not provide high reconstruction accuracy in the spatial domain. To solve this problem, a different structure of 3D-CNN is suggested in this work. At first, different levels of decomposition dependent on the image types, i.e., low-resolution (LR) or high-resolution (HR) training images, are performed on the images by using the non-subsampled contourlet transform. To do this optimally, a similarity criterion is defined to determine the appropriate order of the decomposition levels. The defined similarity criterion is also used to select the appropriate training data from among the decomposed components at the network input, for the corresponding output. Then, according to the sparsity value of the decomposed components of LR and HR images, proper 3D-CNNs are individually designed. Finally, the networks are trained based on types of relations between the LR and HR image components. The proposed method is experimented on three different datasets. The results show superiority of the proposed method with respect to some competitors in terms of several measures.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11760-022-02241-z/MediaObjects/11760_2022_2241_Fig8_HTML.png)
Similar content being viewed by others
References
Zhang, S., Fu, G., Wang, H., et al.: Degradation learning for unsupervised hyperspectral image super-resolution based on generative adversarial network. SIViP 15, 1695–1703 (2021)
He, W., Chen, Y., Yokoya, N., Li, C., Zhao, Q.: Hyperspectral super-resolution via coupled tensor ring factorization. Pattern Recogn. 122, 108280 (2022)
Lu, R., Chen, B., Cheng, Z., Wang, P.: RAFnet: recurrent attention fusion network of hyperspectral and multispectral images. Signal Process. 177, 107737 (2020)
Dian, R., Li, S., Fang, L., Wei, Q.: Multispectral and hyperspectral image fusion with spatial-spectral sparse representation. Inf. Fusion 49, 262–270 (2019)
Chen, H., He, X., Qing, L., Wu, Y., Ren, C., Sheriff, R.E., Zhu, C.: Real-world single image super-resolution: a brief review. Inf. Fusion 79, 124–145 (2022)
Symolon, W., Dagli, C.: Single-image super resolution using convolutional neural network. Procedia Comput. Sci. 185, 213–222 (2021)
Gajbhar, S.S., Joshi, M.V.: Design of complex adaptive multiresolution directional filter bank and application to pansharpening. SIViP 11, 259–266 (2017)
Dian, R., Li, S., Sun, B., Guo, A.: Recent advances and new guidelines on hyperspectral and multispectral image fusion. Inf. Fusion 69, 40–51 (2021)
Imani, M., Ghassemian, H.: Pansharpening optimisation using multiresolution analysis and sparse representation. Int. J. Image Data Fusion 8(3), 270–292 (2017)
Imani, M.: A collaborative representation-based approximation method for remote sensing image fusion. Int. J. Remote Sens. 41(3), 974–995 (2020)
Zhang, L., Nie, J., Wei, W., Li, Y., Zhang, Y.: Deep blind hyperspectral image super-resolution. IEEE Trans. Neural Netw. Learn. Syst. 32(6), 2388–2400 (2021)
Li, J., Cui, R., Li, B., Song, R., Li, Y., Du, Q.: Hyperspectral image super-resolution with 1D–2D attentional convolutional neural network. Remote Sens 11(23), 2859 (2019)
Chen, W., Zheng, X., Lu, X.: Hyperspectral image super-resolution with self-supervised spectral-spatial residual network. Remote Sens 13(7), 1260 (2021)
Patel, R.C., Joshi, M.V.: Super-resolution of hyperspectral images () use of optimum wavelet filter coefficients and sparsity regularization. IEEE Trans. Geosci. Remote Sens. 53(4), 1728–1736 (2015)
Mei, S., Yuan, X., Ji, J., Zhang, Y., Wan, S., Du, Q.: Hyperspectral image spatial super-resolution via 3D full convolutional neural network. Remote Sens. 9(11), 1139 (2017)
Li, Q., Wang, Q., Li, X.: Mixed 2D/3D convolutional network for hyperspectral image super-resolution. Remote Sens. 12(10), 1660 (2020)
Li, Q., Wang, Q., Li X.: Exploring the relationship between 2D/3D convolution for hyperspectral image super-resolution. IEEE Trans. Geosci. Remote Sens. 1–11 (2021)
Wang, L., Bi, T., Shi, Y.: A frequency-separated 3D-CNN for hyperspectral image super-resolution. IEEE Access 8, 86367–86379 (2020)
Yang, J., Zhao, Y.Q., Chan, J.C.W., Xiao, L.: A multi-scale wavelet 3D-CNN for hyperspectral image super-resolution. Remote Sens. 11(13), 1557 (2019)
Cunha, A.L.D., Zhou, J., Do, M.N.: The nonsubsampled contourlet transform: theory, design, and applications. IEEE Trans. Image Process. 15(10), 3089–3101 (2006)
Kumar, N., Verma, R., Sethi, A.: Convolutional neural networks for wavelet domain super resolution. Pattern Recogn. Lett. 90, 65–71 (2017)
Dong, C., Loy, C., He, K., Tang, X.: Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 38(2), 295–307 (2016)
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
Farajzadeh, A., Mohammadi, S. & Imani, M. Hyperspectral image super-resolution using multi-scale decomposition and convolutional neural network based on relation type between low- and high-resolution images. SIViP 17, 361–369 (2023). https://doi.org/10.1007/s11760-022-02241-z
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-022-02241-z