EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Modal Multi-Scale Mamba with Frequency and Semantics-Guided Spectral Modeling for Hyperspectral Image Classification

Hui Peng, Wenzhun Huang, Kaihong Chen, Zhongyan Zhou, Zhen Wang

Remote Sensing · 05.07.2026 · 18 (13) · 2207

DOI: 10.3390/rs18132207 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 05.07.2026. Verfasst von Hui Peng und 4 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Hyperspectral image (HSI) classification remains a challenging task due to high spectral redundancy, complex spectral–spatial correlations, and the limited availability of labeled samples. To address these issues, this paper proposes a novel framework termed M3-Mamba, which integrates language-guided, multi-level, and Mamba-based spectral modeling for hyperspectral image classification. The proposed M3-Mamba leverages high-level semantic priors derived from multimodal representations to guide discriminative spectral modeling, enabling effective interaction between semantic information and fine-grained spectral features. In addition, a frequency-aware Mamba-based state space module is introduced to efficiently capture long-range spectral dependencies while avoiding the quadratic computational complexity of conventional attention mechanisms. Meanwhile, a text-guided modulation strategy is designed to adaptively reweight spectral responses under semantic guidance, suppressing redundant or noisy bands and enhancing class-relevant spectral responses without compromising spectral fidelity. This semantic-to-spectral modulation allows M3-Mamba to better cope with spectral variability and inter-class confusion. Extensive experiments conducted on four widely used benchmark datasets, including Indian Pines, Pavia University, Salinas, and Houston datasets, demonstrate thatM3-Mamba achieves competitive overall accuracy, average accuracy, and Kappa coefficient under the adopted benchmark settings. Ablation studies further validate the effectiveness of each key component, confirming that the proposed framework demonstrates promising effectiveness for hyperspectral image classification.

Zitieren

APA

Peng, H., Huang, W., Chen, K., Zhou, Z., & Wang, Z. (2026). Multi-Modal Multi-Scale Mamba with Frequency and Semantics-Guided Spectral Modeling for Hyperspectral Image Classification. Remote Sensing, 18(13), 2207. https://doi.org/10.3390/rs18132207

Vancouver

Peng H, Huang W, Chen K, Zhou Z, Wang Z. Multi-Modal Multi-Scale Mamba with Frequency and Semantics-Guided Spectral Modeling for Hyperspectral Image Classification. Remote Sensing. 2026;18(13):2207. doi:10.3390/rs18132207

BibTeX
@article{Peng2026,
  author = {Peng, Hui and Huang, Wenzhun and Chen, Kaihong and Zhou, Zhongyan and Wang, Zhen},
  title = {Multi-Modal Multi-Scale Mamba with Frequency and Semantics-Guided Spectral Modeling for Hyperspectral Image Classification},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {13},
  pages = {2207},
  doi = {10.3390/rs18132207}
}
Kostenlos starten →