Abstract
Semi-supervised Graph Neural Networks (GNNs), as an effective data representation learning framework, have been applied to hyperspectral image (HSI) classification. Although GNNs can quickly capture the structural information of HSIs, traditional GNN-based methods use the entire graph as input, which consumes huge computational resources. Moreover, using a single form of GNN variant cannot effectively extract the correlation between two different dimensions, leading to insufficient improvement in the impact of noise and spectral differences in HSIs. To address these issues, a Dual-Stream GNN Fusion Network (named DGFNet) is proposed in this paper. This end-to-end network uses subcubes as input to reduce computational costs. The spatial branch uses Graph Attention Network (GAT) to capture the inherent relationships within the HSIs, and the risk of overfitting is reduced by combining a novel graph pooling and local guidance module to preserve important features in subcubes. The spectral branch weights the correlations among bands to obtain more reasonable spectral features. Finally, these spatial, spectral, and structural features are fused and classified using linear layers. Extensive comparative experiments on four benchmark datasets have demonstrated that our model achieves competitive performance, and its low computational cost advantage makes it more feasible and practical compared to most GNN-based methods.
Similar content being viewed by others
Data availability and access
The datasets generated and analysed in the current study comprise four publicly available datasets: Indian Pines, University of Pavia, Houston 2013, and Salinas datasets
References
Ren J, Shi M, Chen J, Wang R, Wang X (2023) Hyperspectral image classification using multi-level features fusion capsule network with a dense structure. Appl Intell 53(11):14162–14181
Wang M, Yan Z, Luo J, Ye Z, He P (2021) A band selection approach based on wavelet support vector machine ensemble model and membrane whale optimization algorithm for hyperspectral image. Appl Intell 1–15
Tsai M-H, Chang Y-L, Shiau J-S, Wang S-M (2020) Exploring the effects of a serious game-based learning package for disaster prevention education: The case of battle of flooding protection. Int J Disaster Risk Reduction 43:101393
Poland TM, Rassati D (2019) Improved biosecurity surveillance of non-native forest insects: a review of current methods. J Pest Sci 92(1):37–49
Syed-Ab-Rahman SF, Hesamian MH, Prasad M (2022) Citrus disease detection and classification using end-to-end anchor-based deep learning model. Appl Intell 52(1):927–938
Zhang J, Zou X, Kuang L-D, Wang J, Sherratt RS, Yu X (2022) Cctsdb 2021: a more comprehensive traffic sign detection benchmark. Hum-centric Comput Inf Sci 12
Mounika K, Aravind K, Yamini M, Navyasri P, Dash S, Suryanarayana V (2021) Hyperspectral image classification using svm with pca. In: 2021 6th International Conference on Signal Processing, Computing and Control (ISPCC), pp. 470–475. IEEE
Li J, Zhu Q (2020) A boosting self-training framework based on instance generation with natural neighbors for k nearest neighbor. Appl Intell 50(11):3535–3553
Bukht TFN, Rahman H, Jalal A (2023) A novel framework for human action recognition based on features fusion and decision tree. In: 2023 4th International Conference on Advancements in Computational Sciences (ICACS), pp. 1–6. IEEE
Ma KY, Chang C-I (2021) Kernel-based constrained energy minimization for hyperspectral mixed pixel classification. IEEE Trans Geosci Remote Sens 60:1–23
Tan X, Gao K, Liu B, Fu Y, Kang L (2021) Deep global-local transformer network combined with extended morphological profiles for hyperspectral image classification. Journal of Applied Remote Sensing 15(3):038509–038509
Hong D, Wu X, Ghamisi P, Chanussot J, Yokoya N, Zhu XX (2020) Invariant attribute profiles: A spatial-frequency joint feature extractor for hyperspectral image classification. IEEE Trans Geosci Remote Sens 58(6):3791–3808
Huang K-K, Ren C-X, Liu H, Lai Z-R, Yu Y-F, Dai D-Q (2021) Hyperspectral image classification via discriminant gabor ensemble filter. IEEE Trans Cybernet 52(8):8352–8365
Vaddi R, Manoharan P (2020) Hyperspectral image classification using cnn with spectral and spatial features integration. Infrared Phys Technol 107:103296
Huang L, Chen Y (2020) Dual-path siamese cnn for hyperspectral image classification with limited training samples. IEEE Geosci Remote Sens Lett 18(3):518–522
Swain S, Banerjee A (2021) Evaluation of dimensionality reduction techniques on hybrid cnn-based hsi classification. Arabian J Geosci 14(24):2806
Chen Y, Xia R, Zou K, Yang K (2023) Rnon: image inpainting via repair network and optimization network. Int J Mach Learn Cybernet 1–17
Chen Y, Xia R, Zou K, Yang K (2023) Ffti: Image inpainting algorithm via features fusion and two-steps inpainting. J Visual Commun Image Representat 91:103776
Zhou W, Kamata S-i, Wang H, Xue X (2023) Multiscanning-based rnn-transformer for hyperspectral image classification. IEEE Trans Geosci Remote Sens
Zhao J, Hu L, Dong Y, Huang L, Weng S, Zhang D (2021) A combination method of stacked autoencoder and 3d deep residual network for hyperspectral image classification. Int J Appl Earth Observation Geoinf 102:102459
Gao H, Yang Y, Lei S, Li C, Zhou H, Qu X (2019) Multi-branch fusion network for hyperspectral image classification. Knowl-Based Syst 167:11–25
Zhang J, Huang H, Jin X, Kuang L-D, Zhang J (2023) Siamese visual tracking based on criss-cross attention and improved head network. Multimed Tools Appl 1–27
Jia S, Lin Z, Xu M, Huang Q, Zhou J, Jia X, Li Q (2020) A lightweight convolutional neural network for hyperspectral image classification. IEEE Trans Geosci Remote Sens 59(5):4150–4163
Han Z, Hong D, Gao L, Zhang B, Huang M, Chanussot J (2022) Autonas: Automatic neural architecture search for hyperspectral unmixing. IEEE Trans Geosci Remote Sens 60:1–14
Zhang J, Zheng Z, Xie X, Gui Y, Kim G-J (2022) Reyolo: A traffic sign detector based on network reparameterization and features adaptive weighting. J Ambient Intell Smart Environs (Preprint) 1–18
Wang W, Dou S, Jiang Z, Sun L (2018) A fast dense spectral-spatial convolution network framework for hyperspectral images classification. Remote Sens 10(7):1068
Wu H, Li D, Wang Y, Li X, Kong F, Wang Q (2021) Hyperspectral image classification based on two-branch spectral-spatial-feature attention network. Remote Sens 13(21):4262
Chen Y, Xia R, Yang K, Zou K (2023) Mffn: Image super-resolution via multi-level features fusion network. Visual Computer 1–16
Chen Y, Xia R, Yang K, Zou K (2023) Dargs: Image inpainting algorithm via deep attention residuals group and semantics. J King Saud Univ-Computer Inf Sci 35(6):101567
Zhang X, Chen S, Zhu P, Tang X, Feng J, Jiao L (2022) Spatial pooling graph convolutional network for hyperspectral image classification. IEEE Trans Geosci Remote Sens 60:1–15. https://doi.org/10.1109/TGRS.2022.3140353
Bai J, Ding B, Xiao Z, Jiao L, Chen H, Regan AC (2021) Hyperspectral image classification based on deep attention graph convolutional network. IEEE Trans Geosci Remote Sens 60:1–16
Guo F, Li Z, Xin Z, Zhu X, Wang L (2021) Zhang J (2021) Dual graph u-nets for hyperspectral image classification. IEEE J Selected Topics Appl Earth Observations Remote Sens 14:8160–8170
Zhang M, Luo H, Song W, Mei H, Su C (2021) Spectral-spatial offset graph convolutional networks for hyperspectral image classification. Remote Sens 13(21):4342
Qin A, Shang Z, Tian J, Wang Y, Zhang T, Tang YY (2018) Spectral-spatial graph convolutional networks for semisupervised hyperspectral image classification. IEEE Geosci Remote Sens Lett 16(2):241–245
Wan S, Gong C, Zhong P, Du B, Zhang L, Yang J (2019) Multiscale dynamic graph convolutional network for hyperspectral image classification. IEEE Trans Geosci Remote Sens 58(5):3162–3177
Liu Q, Xiao L, Yang J, Wei Z (2020) Cnn-enhanced graph convolutional network with pixel-and superpixel-level feature fusion for hyperspectral image classification. IEEE Trans Geosci Remote Sens 59(10):8657–8671
Sha A, Wang B, Wu X, Zhang L (2020) Semisupervised classification for hyperspectral images using graph attention networks. IEEE Geosci Remote Sens Lett 18(1):157–161
Wang T, Wang G, Tan KE, Tan D (2020) Spectral pyramid graph attention network for hyperspectral image classification. arXiv:2001.07108
Ding Y, Zhao X, Zhang Z, Cai W, Yang N (2021) Multiscale graph sample and aggregate network with context-aware learning for hyperspectral image classification. IEEE J Selected Topics Appl Earth Observations and Remote Sens 14:4561–4572
Zhao Z, Wang H, Yu X (2022) Spectral-spatial graph attention network for semisupervised hyperspectral image classification. IEEE Geosci Remote Sens Lett 19
Guo W, Xu G, Liu W, Liu B, Wang Y (2021) Cnn-combined graph residual network with multilevel feature fusion for hyperspectral image classification. IET Computer Vision 15(8):592–607
Dong Y, Liu Q, Du B, Zhang L (2022) Weighted feature fusion of convolutional neural network and graph attention network for hyperspectral image classification. IEEE Trans Image Process 31:1559–1572
Lu Z, Liang S, Yang Q, Du B (2022) Evolving block-based convolutional neural network for hyperspectral image classification. IEEE Trans Geosci Remote Sens 60:1–21
Qu L, Zhu X, Zheng J, Zou L (2021) Triple-attention-based parallel network for hyperspectral image classification. Remote Sens 13(2):324
Maini DS, Aggarwal DAK (2018) Camera position estimation using 2d image dataset
Bhardwaj VK, Maini S (2021) Improving snr of sm-ofi signal using rls adaptive filter algorithm. In: 2021 Sixth International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), pp. 183–187. IEEE
Ye Z, Li H, Song Y, Benediktsson JA, Tang YY (2016) Hyperspectral image classification using principal components-based smooth ordering and multiple 1-d interpolation. IEEE Trans Geosci Remote Sens 55(2):1199–1209
Veličković, P, Cucurull G, Casanova A, Romero A, Lio P, Bengio Y (2017) Graph attention networks. arXiv:1710.10903
Wang D, Du B, Zhang L (2021) Spectral-spatial graph reasoning network for hyperspectral image classification. arXiv:2106.13952
Chauhan S, Singh M, Aggarwal AK (2023) Designing of optimal digital iir filter in the multi-objective framework using an evolutionary algorithm. Eng Appl Artif Intell 119:105803
Chauhan S, Singh M, Aggarwal AK (2023) Investigative analysis of different mutation on diversity-driven multi-parent evolutionary algorithm and its application in area coverage optimization of wsn. Soft Comput 1–27
Roy SK, Manna S, Song T, Bruzzone L (2021) Attention-based adaptive spec-tral-spatial kernel resnet for hyperspectral image classification. IEEE Trans Geosci Remote Sens 59(9):7831–7843. https://doi.org/10.1109/TGRS.2020.3043267
Roy SK, Krishna G, Dubey SR, Chaudhuri BB (2019) Hybridsn: Exploring 3-d-2-d cnn feature hierarchy for hyperspectral image classification. IEEE Geosci Remote Sens Lett 17(2):277–281
Sun L, Zhao G, Zheng Y, Wu Z (2022) Spectral-spatial feature tokenization transformer for hyperspectral image classification. IEEE Trans Geosci Remote Sens 60:1–14
Ma W, Yang Q, Wu Y, Zhao W, Zhang X (2019) Double-branch multi-attention mechanism network for hyperspectral image classification. Remote Sens 11(11):1307
Zhong Z, Li Y, Ma L, Li J, Zheng W-S (2021) Spectral-spatial transformer network for hyperspectral image classification: A factorized architecture search framework. IEEE Trans Geosci Remote Sens 60:1–15
Li R, Zheng S, Duan C, Yang Y, Wang X (2020) Classification of hyperspectral image based on double-branch dual-attention mechanism network. Remote Sens 12(3):582
Liu Q, Dong Y, Zhang Y, Luo H (2022) A fast dynamic graph convolutional network and cnn parallel network for hyperspectral image classification. IEEE Trans Geosci Remote Sens 60:1–15. https://doi.org/10.1109/TGRS.2022.3179419
Xu Y, Li Z, Li W, Du Q, Liu C, Fang Z, Zhai L (2021) Dual-channel residual network for hyperspectral image classification with noisy labels. IEEE Trans Geosci Remote Sens 60:1–11
Zhang X, Shang S, Tang X, Feng J, Jiao L (2022) Spectral partitioning residual network with spatial attention mechanism for hyperspectral image classification. IEEE Trans Geosci Remote Sens 60:1–14. https://doi.org/10.1109/TGRS.2021.3074196
Maaten L, Hinton G (2008) Visualizing data using t-sne. J Mach Learn Res 9(11)
Acknowledgements
The authors would like to thank the anonymous reviewers for their detailed and constructive comments and suggestions, and thank Assoc. Prof. Cong’an Xu of NAU for their constructive suggestions. Thank Assoc. Prof. Shuaishuai FAN of STBU and Prof. Hongyang Bai of NJUST for the numerical calculations in this article
Funding
This research was supported by the National Natural Science Foundation of China (NSFC), grant number 62271499 and U2031138
Author information
Authors and Affiliations
Contributions
Qikang Liu: Data curation, Methodology, Writing-Original draft preparation. Weiming Li: Conceptualization of this study, Software. Shuaishuai Fan: Data curation, Methodology. Cong’an Xu and Hongyang Bai: Writing-Original draft preparation
Corresponding author
Ethics declarations
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper
Ethical and informed consent for data
All authors are informed and consent to publication
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Weiming Li, Qikang Liu, Shuaishuai Fan, Cong’an Xu, and Hongyang Bai these authors contributed equally to this work.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Li, W., Liu, Q., Fan, S. et al. Dual-stream GNN fusion network for hyperspectral classification. Appl Intell 53, 26542–26567 (2023). https://doi.org/10.1007/s10489-023-04960-3
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-023-04960-3