MoEMamba: Mixture-of-Scale-And-Attention-Experts Mamba Network for Remote Sensing Change Detection
Remote Sensing · 03.11.2026 · 18 (21) · 3778
DOI: 10.3390/rs18213778 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 03.11.2026. Verfasst von Pengfei Zhang und 7 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hebei Academy of Sciences, Hebei Natural Science Foundation, Major Science and Technology Program of Hebei Province.
Abstract (Originalsprache)
The Mamba network has recently emerged as a powerful architecture for remote sensing change detection (RSCD). Notably, the fusion of multi-scale features and heterogeneous attention mechanisms has proven effective in enhancing the performance of Mamba network. However, existing methods fail to quantitatively evaluate the actual contribution weights of different components during feature fusion, thereby limiting the interpretability of their improvements. To address this issue, this paper proposes a novel mixture-of-scale-and-attention-experts Mamba Network (MoEMamba) for the RSCD task, which has the following innovations and contributions: (1) a Siamese Mamba architecture is employed to extract features from bi-temporal images; (2) a mixture-of-scale-experts module is designed to enhance the generalization ability across multi-scale objects, which quantitatively evaluates the contribution of multiple scale-specific experts; and (3) a mixture-of-attention-experts module is proposed to improve the accuracy, which adaptively fuses complementary responses of heterogeneous attention mechanisms based on the learned weights. Extensive experiments on the WHU-CD and LEVIR-CD datasets demonstrate that MoEMamba outperforms existing state-of-the-art methods. Specifically, it achieves F1 scores of 94.20% and 92.62%, and IoU values of 89.04% and 86.25%, respectively. Importantly, this study provides valuable insights into the interpretability of multi-scale feature fusion and heterogeneous attention fusion for RSCD research.
Förderung
- Hebei Academy of Sciences
- Hebei Natural Science Foundation
- Major Science and Technology Program of Hebei Province
Zitieren
APA
Zhang, P., Sun, L., Li, C., Jia, X., Wang, Q., Hao, Q., Zuo, L., & Lu, J. (2026). MoEMamba: Mixture-of-Scale-And-Attention-Experts Mamba Network for Remote Sensing Change Detection. Remote Sensing, 18(21), 3778. https://doi.org/10.3390/rs18213778
Vancouver
Zhang P, Sun L, Li C, Jia X, Wang Q, Hao Q, et al. MoEMamba: Mixture-of-Scale-And-Attention-Experts Mamba Network for Remote Sensing Change Detection. Remote Sensing. 2026;18(21):3778. doi:10.3390/rs18213778
BibTeX
@article{Zhang2026,
author = {Zhang, Pengfei and Sun, Leigang and Li, Chang and Jia, Xiaofeng and Wang, Qinyi and Hao, Qingtao and Zuo, Lu and Lu, Junjing},
title = {MoEMamba: Mixture-of-Scale-And-Attention-Experts Mamba Network for Remote Sensing Change Detection},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {21},
pages = {3778},
doi = {10.3390/rs18213778}
}
EUVIMED