Cross dimensional deep fusion learning for an efficient hyperspectral image classification
*S. Tahseen BanuCorresponding authortahseen511@gmail.comDepartment of Computer Science & EngineeringJawaharlal Nehru Technological University AnantapurAnantapur, Andhra Pradesh, 515002, IndiaView full profile → , K. Subba Reddymrsubbareddy@yahoo.comDepartment of Computer Science & EngineeringRajeev Gandhi Memorial College of Engineering & TechnologyNandyal, Andhra Pradesh, 518501, IndiaView full profile →
* Corresponding author · click or hover a name for details
- Received:
- 01 Mar 2026
- Published Online:
- 31 Oct 2026
- Article type:
- Research Article
- Language:
- EN
- Article no.:
- JIOS-2484
- Pages:
- 3565–3574
Abstract
Hyperspectral image (HSI) classification is an emerging need in remote sensing applications. Extraction of spatial and spectral cues, and their relevance, are crucial for the classification of HSI. Notable research has progressed to address challenges in HSI classifications, including pixel boundary correspondences, spectral band similarities, and high-dimensional issues. Recent methods, specifically SclusterFormer, have primarily focused on these issues and have achieved significant improvements in HSI classification. It effectively enables spectral and spatial information using cluster-based attention and processes features independently and combines them through a simple additive fusion strategy, which limits its ability to model cross-dimensional dependencies. To address these limitations, a Cross-Dimensional Spectral–Spatial Fusion Framework (CDSSF) is proposed. The proposed method introduces an adaptive attention-based fusion mechanism that dynamically integrates spectral and spatial features, enabling effective interaction across dimensions. This enhances feature discrimination, improves robustness to spectral variability, and leads to superior classification performance compared to the baseline model.
Keywords
Subject Classifications
References
[1] H. Zhang, K. Tu, H. Lv, and R. Wang, “Hyperspectral image classification based on 3D–2D hybrid convolution and graph attention mechanism,” Neural Process. Lett., vol. 56, Art. no. 117 (2024).
[2] T. Arshad, B. Peng, A. Rahman, R. Khan, S. U. Khan, S. Alnazi, and N. Alturki, “Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification,” Sci. Rep., vol. 15, Art. no. 27259 (2025), doi: 10.1038/s41598-025-12489-3.
[3] Y. Peng, Y. Zhang, B. Tu, Q. Li, and W. Li, “Spatial–spectral transformer with cross-attention for hyperspectral image classification,” IEEE Trans. Geosci. Remote Sens., vol. 60, pp. 1–15 (2022).
[4] Z. Chen, C. Liu, and J. Zhou, “SSIT: A spatial–spectral interactive transformer for hyperspectral image denoising,” Sci. Remote Sens., vol. 12, Art. no. 100276 (2025), doi: 10.1016/j.srs.2025.100276.
[5] K. Ding, T. Lu, W. Fu, S. Li, and F. Ma, “Global–local transformer network for HSI and LiDAR data joint classification,” IEEE Trans. Geosci. Remote Sens., vol. 60, pp. 1–13 (2022), doi: 10.1109/TGRS.2022.3216319.
[6] B. Zhang, Y. Chen, Y. Rong, S. Xiong, and X. Lu, “MATNet: A combining multi-attention and transformer network for hyperspectral image classification,” IEEE Trans. Geosci. Remote Sens., vol. 61, pp. 1–15 (2023).
[7] S. Cheng, Z. Xue, Z. Li, A. Xu, and H. Su, “Spectral–spatial score fusion attention network for hyperspectral image classification with limited samples,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 17, pp. 14521–14542 (2024), doi: 10.1109/JSTARS.2024.3440254.
[8] E. Zhao, Y. Su, N. Qu, Y. Wang, and J. Zeng, “Self- and cross-attention enhanced transformer for visible and thermal infrared hyperspectral image classification,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 18, pp. 13408–13422 (2025), doi: 10.1109/JSTARS.2025.3571226.
[9] Y. He, B. Xi, G. Li, T. Zheng, Y. Li, C. Xue, and J. Chanussot, “Multilevel attention dynamic-scale network for HSI and LiDAR data fusion classification,” IEEE Trans. Geosci. Remote Sens., vol. 62, Art. no. 5529916 (2024), doi: 10.1109/TGRS.2024.3456754.
[10] M. Wang, F. Gao, J. Dong, H.-C. Li, and Q. Du, “Nearest neighbor-based contrastive learning for hyperspectral and LiDAR data classification,” IEEE Trans. Geosci. Remote Sens., vol. 61, Art. no. 5501816 (2023).
[11] Z.-Z. Ge, Z. Ding, Y. Wang, L.-F. Bian, and C. Yang, “Spectral domain strategies for hyperspectral super-resolution: Transfer learning and channel enhance network,” Int. J. Appl. Earth Obs. Geoinf., vol. 134, Art. no. 104180 (2024), doi: 10.1016/j.jag.2024.104180.
[12] M. Zhao, W. Zheng, and J. Hu, “HTC-HAD: A hybrid transformer-CNN approach for hyperspectral anomaly detection,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 18, pp. 10144–10156 (2025).
[13] M. Suleman Basha, S. K. Mouleeswaran, and K. Rajendra Prasad, “Hybrid visual computing models to discover the clusters assessment of high dimensional big data,” Soft Comput., vol. 27, pp. 4249–4262 (2023).
[14] Y. Zhong, X. Hu, C. Luo, X. Wang, J. Zhao, and L. Zhang, “WHU-Hi: UAV-borne hyperspectral with high spatial resolution (H²) benchmark datasets and classifier for precise crop identification based on deep convolutional neural network with CRF,” Remote Sens. Environ., vol. 250, Art. no. 112012 (2020), doi: 10.1016/j.rse.2020.112012.
[15] U. Samantsinghar, S. Sethi, R. C. Sahoo, and D. C. Panda, “Optimizing spectrum sensing performance and prediction of trustworthy users by implementing machine learning algorithms in cognitive radio network,” J. Inf. Optim. Sci., vol. 46, no. 4-B, pp. 1369–1383 (2025).
[16] H. Yu, H. Yang, L. Gao, J. Hu, A. Plaza, and B. Zhang, “Hyperspectral image change detection based on gated spectral–spatial–temporal attention network with spectral similarity filtering,” IEEE Trans. Geosci. Remote Sens., vol. 62, pp. 1–13 (2024), doi: 10.1109/TGRS.2024.3373820.
[17] B. Gayathri and R. Gopi, “Some properties of k-even mean graph,” Journal of Discrete Mathematical Sciences and Cryptography, vol. 29, no. 4, pp. 1579–1583 (2025), doi: 10.47974/JDMSC-2244.
[18] M. Vasti and A. Dev, “Ensemble learning based predictive modelling on a highly imbalanced multiclass data,” J. Inf. Optim. Sci., vol. 45, no. 8, pp. 2141–2164 (2024), doi: 10.47974/JIOS-1778.
