H2O-Net: Self-Supervised Flood Segmentation via Adversarial Domain Adaptation and Label Refinement
This addresses the problem of unreliable flood alerting due to low-resolution, low-latency satellite data for disaster response, though it is incremental in improving segmentation accuracy.
The paper tackles flood detection from satellite and aerial imagery by introducing H2O-Net, a self-supervised method using adversarial domain adaptation and label refinement, which outperforms state-of-the-art semantic segmentation methods by 10% pixel accuracy and 12% mIoU.
Accurate flood detection in near real time via high resolution, high latency satellite imagery is essential to prevent loss of lives by providing quick and actionable information. Instruments and sensors useful for flood detection are only available in low resolution, low latency satellites with region re-visit periods of up to 16 days, making flood alerting systems that use such satellites unreliable. This work presents H2O-Network, a self supervised deep learning method to segment floods from satellites and aerial imagery by bridging domain gap between low and high latency satellite and coarse-to-fine label refinement. H2O-Net learns to synthesize signals highly correlative with water presence as a domain adaptation step for semantic segmentation in high resolution satellite imagery. Our work also proposes a self-supervision mechanism, which does not require any hand annotation, used during training to generate high quality ground truth data. We demonstrate that H2O-Net outperforms the state-of-the-art semantic segmentation methods on satellite imagery by 10% and 12% pixel accuracy and mIoU respectively for the task of flood segmentation. We emphasize the generalizability of our model by transferring model weights trained on satellite imagery to drone imagery, a highly different sensor and domain.