Gao et al. (2025) MFF-Net: Flood Detection from SAR Images Using Multi-Frequency and Fuzzy Uncertainty Fusion

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Remote Sensing
• Year: 2025
• Date: 2025-12-29
• Authors: Yahui Gao, X. Renshaw Wang, Zili Zhang, Xu Chen, Ruijun Liu, Xiaohui Liang
• DOI: 10.3390/rs18010123

Research Groups

Not specified in the provided text.

Short Summary

This study proposes MFF-Net, a novel multi-scale deep learning algorithm, to overcome systematic noise in Synthetic Aperture Radar (SAR) images, significantly improving pixel-level and fine-grained flood detection accuracy.

Objective

• To develop a novel flood detection algorithm (MFF-Net) that mitigates systematic noise in Synthetic Aperture Radar (SAR) images and improves the accuracy of pixel-level and fine-grained flood area detection.

Study Configuration

• Spatial Scale: Pixel-level, focusing on fine-grained flood areas.
• Temporal Scale: Event-based, tested on actual flood events.

Methodology and Data

• Models used: Multi-frequency fuzzy uncertainty fusion network (MFF-Net).
• Data sources: Synthetic Aperture Radar (SAR) images; specifically, MMflood Dataset, Sen1Floods11 Dataset, ETCI 2021 Dataset, and SAR Poyang Lake Water Body Sample Dataset.

Main Results

• MFF-Net achieved Intersection over Union (IoU) scores of 50.2% on the MMflood Dataset, 45.07% on the Sen1Floods11 Dataset, 44.35% on the ETCI 2021 Dataset, and 57.27% on the SAR Poyang Lake Water Body Sample Dataset.
• The algorithm demonstrated enhanced detection capability for fine-grained flood areas and was validated on actual flood events.

Contributions

• Introduction of MFF-Net, a novel multi-scale flood detection algorithm that leverages multi-frequency feature extraction and fuzzy uncertainty fusion modules to effectively mitigate systematic noise in SAR images.
• Significant improvement in the accuracy of pixel-level and fine-grained flood detection compared to existing methods.

Funding

Not specified in the provided text.

Citation

@article{Gao2025MFFNet,
  author = {Gao, Yahui and Wang, X. Renshaw and Zhang, Zili and Chen, Xu and Liu, Ruijun and Liang, Xiaohui},
  title = {MFF-Net: Flood Detection from SAR Images Using Multi-Frequency and Fuzzy Uncertainty Fusion},
  journal = {Remote Sensing},
  year = {2025},
  doi = {10.3390/rs18010123},
  url = {https://doi.org/10.3390/rs18010123}
}