A Spatially Masked Adaptive Gated Network for multimodal post-flood water extent mapping using SAR and incomplete multispectral data
Hyunho Lee, Wenwen Li
TL;DR
This work tackles post-flood water extent mapping by leveraging SAR data and partially available MSI through a novel Spatially Masked Adaptive Gated Network (SMAGNet). SMAGNet uses dual-stream encoders, a Spatially Masked Adaptive Gated Feature Fusion Module (SMAG-FFM) with a spatial mask to handle missing MSI pixels, and a weight-shared decoder to jointly learn SAR-only and SAR–MSI fused representations. Across the C2S-MS Floods dataset, SMAGNet achieves superior IoU and recall while maintaining robustness as MSI data becomes increasingly incomplete, and it remains competitive with SAR-only baselines when MSI are entirely missing. The model also demonstrates strong generalizability to a real-world flood event with substantial MSI gaps, indicating practical applicability for disaster response. Limitations include performance in densely vegetated areas and the need for more diverse benchmark datasets for multimodal post-flood mapping.
Abstract
Mapping water extent during a flood event is essential for effective disaster management throughout all phases: mitigation, preparedness, response, and recovery. In particular, during the response stage, when timely and accurate information is important, Synthetic Aperture Radar (SAR) data are primarily employed to produce water extent maps. Recently, leveraging the complementary characteristics of SAR and MSI data through a multimodal approach has emerged as a promising strategy for advancing water extent mapping using deep learning models. This approach is particularly beneficial when timely post-flood observations, acquired during or shortly after the flood peak, are limited, as it enables the use of all available imagery for more accurate post-flood water extent mapping. However, the adaptive integration of partially available MSI data into the SAR-based post-flood water extent mapping process remains underexplored. To bridge this research gap, we propose the Spatially Masked Adaptive Gated Network (SMAGNet), a multimodal deep learning model that utilizes SAR data as the primary input for post-flood water extent mapping and integrates complementary MSI data through feature fusion. In experiments on the C2S-MS Floods dataset, SMAGNet consistently outperformed other multimodal deep learning models in prediction performance across varying levels of MSI data availability. Furthermore, we found that even when MSI data were completely missing, the performance of SMAGNet remained statistically comparable to that of a U-Net model trained solely on SAR data. These findings indicate that SMAGNet enhances the model robustness to missing data as well as the applicability of multimodal deep learning in real-world flood management scenarios.