2026· IEEE Geoscience and Remote Sensing Letters· Vol 23, pp. 1506005-1506005· 0 citations· 18 references
Abstract
Flood detection utilizing synthetic aperture radar (SAR) imagery offers critical support for disaster response and emergency decision-making. Recently, Mamba-based methods have been increasingly adopted in remote sensing change detection (CD). By leveraging its efficient long-sequence modeling capability, the architecture can effectively capture large-scale inundation changes. However, considering the complexity of flood distribution, interference from weak scatterers, and speckle noise severely limits the intelligent detection accuracy of inundated regions in SAR images. To address these challenges, this letter proposes a novel SAR flood detection network named SF-Mamba. Our method employs a visual Mamba architecture as an encoder to effectively extract global contextual features from SAR images. Furthermore, we design a Dual-path Change-aware Fusion Module (DCFM), which captures both differential and semantic change information. By utilizing state space model (SSM) and strip convolution, DCFM improves the characterization of integrity in both large-scale and fragmented inundation areas. Additionally, a learnable parameter is introduced to dynamically fuse two types of change information, effectively suppressing interference caused by weak scatterers and speckle noise. Finally, a multiscale deep supervision strategy is adopted to guide the generation of flood inundation maps. Comparative experiments on two public datasets demonstrate that SF-Mamba achieves state-of-the-art performance. The source code is publicly available at https://github.com/masurq/SF-Mamba
The proposed Mamba-enhanced Real-Time Detection Transformer (RT-DETR) framework for high-precision ship detection and identification achieves strong robustness and efficiency, showing high practical value for maritime monitoring, port security, and emergency response.
This article proposes a dual-regularized SAR-guided fusion network that leverages the structural stability of SAR features to adaptively refine MS feature representations, enabling reliability-aware multimodal fusion.
Yan-Kun Huang, Zhi-Guo Liang, Jiao-Jiao Tian et al.· IEEE Journal of Selected Top...· 0 citations
Flood mapping from predisaster optical and postdisaster synthetic aperture radar (SAR) imagery is important for emergency response, yet severe cross-modal discrepancy often induces pseudo changes in unchanged areas and weakens change evidence in newly inundated regions. To address this issue, we propose SADRNet, an OPT...
Synthetic Aperture Radar (SAR) satellites have emerged as a crucial technology for real-time flood monitoring in the face of escalating flood disaster hazards. However, existing flood extraction techniques still face challenges in simultaneously achieving accurate urban inundation detection and extensive large-scale...
Run Xing, Feng-Yuan Lu, Jun Zhao· Natural Hazards and Earth Sy...· 0 citations
M-FSAD-KD is proposed, a full-link distillation framework whose neck-stage Fourier-gated alignment transfers low-frequency structural content while preserving target-edge high-frequency content; a joint spatial–channel attention mask, a shallow backbone adapter, and a response-level knowledge distillation (KD) term com...
Yu-Ming Tong, Kai-Na Xiong, Jun Liu et al.· Remote Sensing· 0 citations
Synthetic Aperture Radar (SAR), utilizing its ability to actively transmit microwave signals, is widely used for military aircraft target reconnaissance and aviation safety monitoring. However, the complex background of ground-based aircraft targets, combined with the coherent imaging characteristics of SAR, results in...
Qing Guo, Hongdong Zhao, Wen-Jing Wang et al.· Radioengineering· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.