EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MMDFRNet: Dynamic Cross-Modal Decoupling and Alignment for Robust Rice Mapping

Tingyan Fu, Jia Ge, Shufang Tian

Remote Sensing · 02.05.2026 · 18 (9) · 1413

DOI: 10.3390/rs18091413 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 02.05.2026. Verfasst von Tingyan Fu und 2 weiteren Autor:innen. Die Arbeit nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Mine Development and Ecological Space Monitoring and Evaluation in Key Areas, China University of Geosciences.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate rice mapping is critical for grain yield estimation and food security, yet traditional methods often struggle with asynchronous data quality and the inherent statistical gap between SAR and optical signals. To bridge this gap, we propose MMDFRNet, a novel multi-modal deep learning framework that synergistically integrates Sentinel-1 SAR and Sentinel-2 optical imagery. Unlike conventional static fusion approaches, MMDFRNet features a dual-stream modality-specific encoder architecture designed to decouple structural backscattering signals from spectral reflectance. Central to this framework is the multi-modal feature fusion (MMF) module, which employs an adaptive attention mechanism to dynamically align and recalibrate features based on their reliability, effectively mitigating noise from compromised modalities. Additionally, a multi-scale feature fusion (MSF) module is incorporated to coordinate hierarchical semantic information, enhancing boundary delineation in fragmented landscapes. Extensive experiments conducted across multiple study areas in China demonstrate the superiority of MMDFRNet. The model achieves a Precision of 0.9234, an IoU of 0.8612, and an F1-score of 0.9252. Notably, it consistently outperforms state-of-the-art benchmarks (e.g., UNetFormer, STMA, and CCRNet) by margins of up to 11.72% (Precision) and 7.39% (IoU) compared to classic baselines. Furthermore, rigorous ablation studies and degradation analyses confirm the model’s robustness, verifying its ability to transform the degradation paradox into a performance booster through pixel-wise adaptive alignment. Consequently, MMDFRNet offers a promising solution for precise rice area statistics and long-term monitoring in complex agricultural landscapes.

Förderung

  • National Mine Development and Ecological Space Monitoring and Evaluation in Key Areas, China University of Geosciences

Zitieren

APA

Fu, T., Ge, J., & Tian, S. (2026). MMDFRNet: Dynamic Cross-Modal Decoupling and Alignment for Robust Rice Mapping. Remote Sensing, 18(9), 1413. https://doi.org/10.3390/rs18091413

Vancouver

Fu T, Ge J, Tian S. MMDFRNet: Dynamic Cross-Modal Decoupling and Alignment for Robust Rice Mapping. Remote Sensing. 2026;18(9):1413. doi:10.3390/rs18091413

BibTeX
@article{Fu2026,
  author = {Fu, Tingyan and Ge, Jia and Tian, Shufang},
  title = {MMDFRNet: Dynamic Cross-Modal Decoupling and Alignment for Robust Rice Mapping},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {9},
  pages = {1413},
  doi = {10.3390/rs18091413}
}
Kostenlos starten →