journal article·2025-05-23Open access

Explainable Two-Layer Mode Machine Learning Method for Hyperspectral Image Classification

Wenjia Chen, Junwei Cheng, Song Yang, Li Sun

Applied Sciences· Volume 15· Issue 11

Abstract

Explainable machine learning methods with a specific mathematical model provide insights into how the model works. We propose a new mode that contains a two-layer architecture for hyperspectral image (HSI) classification. In the front-end learning layer, superpixel segmentation and mathematical models are combined to achieve the band selection, which obtains the data re-expression in a lower dimension. The mathematical model uses the l2,1 norm and graph regularized term, which helps induce sparsity, improve robustness to outliers and noise, and enhance the explainability of the data re-expression. We employ the support vector machine or the K-nearest neighbor algorithms in the back-end layer to classify low-dimensional data. Finally, the two-layer mode classification method is applied to the three real HSI dataset classifications. Numerical results show that the overall classification accuracy of our method is improved.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Remote-Sensing Image Classification · Anomaly Detection Techniques and Applications · Image and Signal Denoising Methods

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

College of Information Science and Engineering, Shandong Agricultural University, Tai’an 271018, China

Postdoctoral Research Station for Agricultural Resource Utilization, Shandong Agricultural University, Tai’an 271018, China

College of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China

Funding

Strategic Priority Research Program of Chinese Academy of Sciences

References (38)

Show source-provided references
1. Cheng, M.-F., Mukundan, A., Karmakar, R., Valappil, M.A.E., Jouhar, J., and Wang, H.C. (2025). Modern Trends and Recent Applications of Hyperspectral Imaging: A Review. Technologies, 13.10.3390/technologies13050170 ↗
2. Zhang, Y., Wu, W., Zhou, X., and Cheng, J.H. (2025). Non-Destructive Detection of Soybean Storage Quality Using Hyperspectral Imaging Technology. Molecules, 30.10.3390/molecules30061357 ↗
3. UAV Hyperspectral Remote Sensing Image Classification: A Systematic Review10.1109/JSTARS.2024.3522318 ↗
4. Pushpalatha, V., Mallikarjuna, P.B., Mahendra, H.N., Subramoniam, S.R., and Mallikarjunaswamy, S. (2025). Land use and land cover classification for change detection studies using convolutional neural network. Appl. Comput. Geosci., 25.10.1016/j.acags.2025.100227 ↗
5. Valme, D., Rassõlkin, A., and Liyanage, D.C. (2025). From ADAS to Material-Informed Inspection: Review of Hyperspectral Imaging Applications on Mobile Ground Robots. Sensor, 25.10.3390/s25082346 ↗
6. Ma, K., Yao, C., Liu, B., Hu, Q., Li, S., He, P., and Han, J. (2025). Segment Anything Model-Based Hyperspectral Image Classification for Small Samples. Remote Sens., 17.10.3390/rs17081349 ↗
7. Hennessy, A., Clarke, K., and Lewis, M. (2020). Hyperspectral Classification of Plants: A Review of Waveband Selection Generalisability. Remote Sens., 12.10.3390/rs12010113 ↗
8. Robust hyperspectral classification using relevance vector machine10.1109/TGRS.2010.2103381 ↗
9. Zhao, X., Ma, J., Wang, L., Zhang, Z., Ding, Y., and Xiao, X. (2025). A review of hyperspectral image classification based on graph neural networks. Artif. Intell. Rev., 58.10.1007/s10462-025-11169-y ↗
10. An enhanced semi-supervised support Vector Machine Algorithm for spectral-spatial hyperspectral image classification10.1134/S1054661824010085 ↗
11. Chen, C., Yuan, X., Gan, S., Kang, X., Luo, W., Li, R., and Gao, S. (2024). A new strategy based on multi-source remote sensing data for improving the accuracy of land use/cover change classification. Sci. Rep., 14.10.1038/s41598-024-75329-w ↗
12. Composite spectral spatial pixel CNN for land-use hyperspectral image classification with hybrid activation function10.1007/s11042-024-19327-0 ↗
13. Liu, J., Lan, J., Zeng, Y., Luo, W., Zhuang, Z., and Zou, J. (2025). Explainability Feature Bands Adaptive Selection for Hyperspectral Image Classification. Remote Sens., 17.10.3390/rs17091620 ↗
14. Interpreting black-box models: A review on explainable artificial intelligence10.1007/s12559-023-10179-8 ↗
15. Explainable artificial intelligence for spectroscopy data: A review10.1007/s00424-024-02997-y ↗
16. Li, R., Zheng, S., Duan, C., Yang, Y., and Wang, X. (2020). Classification of hyperspectral image based on double-branch dual-attention mechanism network. Remote Sens., 12.10.20944/preprints201912.0059.v2 ↗
17. Dual-channel convolution network with image-based global learning framework for hyperspectral image classification
18. Self-supervised learning with a dual-branch ResNet for hyperspectral image classification
19. Wang, C., Zhan, C., Lu, B., Yang, W., Zhang, Y., Wang, G., and Zhao, Z. (2024). SSFAN: A Compact and Efficient Spectral-Spatial Feature Extraction and Attention-Based Neural Network for Hyperspectral Image Classification. Remote Sens., 16.10.3390/rs16224202 ↗
20. Tang, X., Zhang, K., Zhou, X., Zeng, L., and Huang, S. (2024). Enhancing Binary Convolutional Neural Networks for Hyperspectral Image Classification. Remote Sens., 16.10.3390/rs16234398 ↗
21. Zhao, H., Lu, Z., Sun, S., Wang, P., Jia, T., Xie, Y., and Xu, F. (2025). Classification of Large Scale Hyperspectral Remote Sensing Images Based on LS3EU-Net++. Remote Sens., 17.10.3390/rs17050872 ↗
22. Graph regularized spatial–spectral subspace clustering for hyperspectral band selection10.1016/j.neunet.2022.06.016 ↗
23. Wang, Y., Ma, H., Yang, Y., Zhao, E., Song, M., and Yu, C. (2024). Self-supervised deep multi-level representation learning fusion-based maximum entropy subspace clustering for hyperspectral band selection. Remote Sens., 16.10.3390/rs16020224 ↗
24. A fast neighborhood grouping method for hyperspectral band selection10.1109/TGRS.2020.3011002 ↗
25. Liu, K., Chen, Y., and Chen, T. (2022). A band subset selection approach based on sparse self-representation and band grouping for hyperspectral image classification. Remote Sens., 14.10.3390/rs14225686 ↗
26. Spectral-spatial genetic algorithm-based unsupervised band selection for hyperspectral image classification10.1109/TGRS.2020.3047223 ↗
27. Supervised band selection in hyperspectral images using single-layer neural networks10.1080/01431161.2018.1553322 ↗
28. Sparse subspace clustering: Algorithm, theory, and applications10.1109/TPAMI.2013.57 ↗
29. Band selection using improved sparse subspace clustering for hyperspectral imagery classification10.1109/JSTARS.2015.2417156 ↗
30. On spectral clustering: Analysis and an algorithm
31. Fusion of front-end and back-end learning based on layer-by-layer data re-representation10.1360/N112017-00268 ↗
32. Liu, M., Tuzel, O., Ramalingam, S., and Chellappa, R. (2011, January 20–25). Entropy rate superpixel segmentation. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Colorado Springs, CO, USA.10.1109/CVPR.2011.5995323 ↗
33. Superpixel segmentation based on anisotropic diffusion model for object-oriented remote sensing image classification10.1109/JSTARS.2023.3324770 ↗
34. Hyperspectral band selection via adaptive subspace partition strategy10.1109/JSTARS.2019.2941454 ↗
35. A hyperspectral band selection method via adjacent subspace partition
36. Spatial and spectral structure preserved self-representation for unsupervised hyperspectral band selection
37. Hyperspectral band selection via region-aware latent features fusion based clustering10.1016/j.inffus.2021.09.019 ↗
38. Wang, H., Yu, G., Cheng, J., Zhang, Z., Wang, X., and Xu, Y. (2024). Fast hyperspectral image classification with strong noise robustness based on minimum noise fraction. Remote Sens., 16.10.3390/rs16203782 ↗