A comprehensive survey for Hyperspectral Image Classification: The evolution from conventional to transformers and Mamba models

Ahmad, Muhammad; Distefano, Salvatore; Khan, Adil Mehmood; Mazzara, Manuel; Li, Chenyu; Li, Hao; Aryal, Jagannath; Ding, Yao; Vivone, Gemine; Hong, Danfeng

Abstract

Hyperspectral Image Classification (HSIC) presents significant challenges owing to the high dimensionality and intricate nature of Hyperspectral data. While traditional Machine Learning (TML) approaches have demonstrated effectiveness, they often encounter substantial obstacles in real-world applications, including the variability of optimal feature sets, subjectivity in human-driven design, inherent biases, and methodological limitations. Specifically, TML suffers from the curse of dimensionality, difficulties in feature selection and extraction, insufficient consideration of spatial information, limited robustness against noise, scalability issues, and inadequate adaptability to complex data distributions. In recent years, Deep Learning (DL) techniques have emerged as robust solutions to address these challenges. This survey offers a comprehensive overview of current trends and future prospects in HSIC, emphasizing advancements from DL models to the increasing adoption of Transformer and Mamba Model architectures. We systematically review key concepts, methodologies, and state-of-the-art approaches in DL for HSIC. Furthermore, we investigate the potential of Transformer-based models and the Mamba Model in HSIC, detailing their advantages and challenges. Emerging trends in HSIC are explored, including in-depth discussions on Explainable AI and interpretability concepts, alongside Diffusion Models for denoising, feature extraction, and fusion. Comprehensive experimental results were conducted on three Hyperspectral datasets to substantiate the efficacy of various conventional DL models. Additionally, we identify several open challenges and pertinent research questions in the field of HSIC. Finally, we outline future research directions and potential applications aimed at enhancing the accuracy and efficiency of HSIC. The Source code is available at https://github.com/mahmad000/HSIC-2024.

Citation

Ahmad, M., Distefano, S., Khan, A. M., Mazzara, M., Li, C., Li, H., Aryal, J., Ding, Y., Vivone, G., & Hong, D. (2025). A comprehensive survey for Hyperspectral Image Classification: The evolution from conventional to transformers and Mamba models. Neurocomputing, 644, Article 130428. https://doi.org/10.1016/j.neucom.2025.130428

• Journal Article Type: Article
• Acceptance Date: May 2, 2025
• Online Publication Date: May 20, 2025
• Publication Date: Sep 1, 2025
• Deposit Date: May 29, 2025
• Publicly Available Date: May 21, 2026
• Print ISSN: 0925-2312
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 644
• Article Number: 130428
• DOI: https://doi.org/10.1016/j.neucom.2025.130428
• Public URL: https://hull-repository.worktribe.com/output/5184546