Flood Detection

Flood detection research focuses on rapidly and accurately mapping flood extents using remote sensing data, primarily satellite and aerial imagery, to support disaster response and management. Current efforts leverage deep learning models, including convolutional neural networks (CNNs), recurrent neural networks (RNNs like LSTMs), and transformers (like Vision Transformers), often incorporating techniques like semantic segmentation and anomaly detection to analyze both single-time and multi-temporal data (e.g., SAR and multispectral imagery). These advancements improve the speed and accuracy of flood mapping, enabling more effective resource allocation and timely warnings, particularly in conjunction with crowdsourced data and improved data processing tools like Apache Sedona.