Understanding flood detection models across Sentinel-1 and Sentinel-2 modalities and benchmark datasets

In recent years, research in flood mapping from remote sensing satellite imagery has predominantly focused on deep learning methods. While new flood segmentation models are increasingly being proposed, most of these works focus on advancing architectures trained on single datasets. Therefore, these studies overlook the intrinsic limitations and biases of the available training and evaluation data. This often leads to poor generalization and overconfident predictions when these models are used in real-world scenarios. To address this gap, the objective of this work is twofold. First, we train and evaluate flood segmentation models on five publicly available datasets including data from Sentinel-1, Sentinel-2, and both SAR and multispectral modalities. Our findings reveal that models achieving high detection accuracy on a single dataset (intra-dataset validation) do not necessarily generalize well to unseen datasets. In contrast, models trained on more diverse samples from multiple datasets demonstrate greater robustness and generalization ability. Furthermore, we present a dual-stream multimodal architecture that can be independently trained and tested on both single-modality and dual-modality datasets. This enables the integration of all the diversity and richness of the available data into a single unified framework. The results emphasize the need for a more comprehensive validation using diverse and well-designed datasets, particularly for multimodal approaches. If not adequately addressed, the shortcomings of current datasets can significantly limit the potential of deep learning-based operational flood mapping approaches.