Beyond Major Floods: Deep Learning for Detecting Shallow Water Inundation in Agricultural Areas

Flood detection using satellite imagery is crucial for environmental monitoring and disaster management, especially in rural and agricultural regions where even minor water inundation can disrupt farmland accessibility and road safety. Sentinel-1 Synthetic Aperture Radar (SAR) imagery offers a robust solution for mapping water under various weather conditions. Although deep learning-based segmentation methods have shown promising results for food detection, their comparative performance in agricultural landscapes, including small-scale surface water dynamics, remains underexplored. In this study, we introduced a three-class segmentation framework that distinguishes sea, inland water, and land, improving the food detection accuracy in complex coastal farmland. Ten different deep learning models were evaluated for segmentation using Sentinel-1 VH polarization decibel values. We further investigated anomaly detection via autoencoders and variational autoencoders to track temporal changes in food-prone areas. To handle large-scale satellite imagery more effectively, we tiled the images, ensuring that the segmentation models could process high-resolution data efficiently. The evaluations showed that the DeepLabv3+ and hybrid ResNet-UNet models outperformed the others. Despite having a more lightweight architecture and lower computational and memory resource requirements, the ResNet-UNet model achieved predictive performance comparable to that of DeepLabv3+.