🤖 AI Summary
High-resolution flood mapping is often hindered by cloud cover in optical imagery and speckle noise along with georegistration errors in synthetic aperture radar (SAR) data. To address these challenges, this work proposes a multimodal flood mapping framework that integrates Sentinel-1 and Sentinel-2 observations, introducing three key innovations: a translation-invariant loss function robust to registration offsets, a generative SAR despeckling model based on a conditional variational autoencoder (CVAE), and a weakly supervised label transfer strategy. Evaluated on a newly curated, high-quality multisensor flood dataset covering the conterminous United States, the proposed method achieves substantially improved accuracy under complex weather and urban conditions, attaining a multispectral area under the precision–recall curve (AUPRC) of 0.956 and demonstrating markedly superior SAR-based mapping performance compared to conventional filtering approaches.
📝 Abstract
Reliable high-resolution flood extent mapping from satellite imagery remains constrained by limited data fidelity and sensor-specific artifacts. Multispectral optical imagery is degraded by clouds, shadows, and urban confounders, while synthetic aperture radar (SAR) imagery is affected by speckle noise and sensor co-registration uncertainty. This work presents an integrated flood mapping framework that jointly addresses these limitations through curated datasets and novel learning strategies. We introduce a new Sentinel-2 (S2) and Sentinel-1 (S1) dataset covering the contiguous United States, featuring pixel-accurate 10 m water masks with emphasis on challenging weather conditions and urban environments that are underrepresented in existing benchmarks. High-quality S2 annotations are manually produced using rigorous geospatial labeling protocols and transferred to SAR imagery through weakly labeled temporally coincident acquisitions. To address SAR-specific artifacts, a shift-invariant loss function is employed to tolerate residual geolocation uncertainty between SAR imagery and optical-derived labels, and a Conditional Variational Autoencoder (CVAE) is trained on multitemporal SAR composites to suppress speckle while preserving flood-relevant spatial structure. Experiments using UNet and UNet++ architectures demonstrate strong multispectral performance (AUPRC up to 0.956) and statistically significant improvements in SAR flood mapping when using shift-invariant loss and CVAE-based despeckling compared to classical filters. These results underscore the importance of dataset fidelity, misalignment-robust training, and demonstrate the viability of generative despeckling for operational flood mapping.