Home/Datasets/ANRF-MAHA-AISEHack-Apr2026-Phase2-Theme-1-Flood-Prediction

ANRF-MAHA-AISEHack-Apr2026-Phase2-Theme-1-Flood-Prediction

AISEHack 2026 (ANRF, with IBM Research) multisensor flood prediction dataset using EOS-4 SAR (HH, HV) and Resourcesat optical data with co-registered 512×512 patches and 3-class labels.

About Dataset

This dataset has been curated as part of AISEHack 2026 (Hackathon #1), an initiative under the ANRF AI for Science program, in collaboration with IBM Research. The hackathon focuses on advancing AI-driven solutions for real-world scientific challenges, with Theme #1 centered on flood prediction and monitoring using remote sensing data. The dataset consists of high-resolution multisensor satellite imagery acquired from two indigenous Indian platforms: EOS-4 and Resourcesat-2/2A. EOS-4 provides S

Purpose of Dataset

This Dataset Is Designed To Support Research And Development In Flood Prediction And Monitoring Using Multimodal Remote Sensing Data. It Enables The Development And Benchmarking Of Machine Learning And Deep Learning Models For Semantic Segmentation Tasks, Particularly In Disaster Management And Climate Resilience Applications. By Combining Sar And Optical Imagery With High-quality Ground Truth Labels, The Dataset Provides A Robust Foundation For Building Models That Can Operate Under Diverse Environmental Conditions, Including Cloud Cover And Low-visibility Scenarios. It Also Serves As A Standardized Benchmark For Participants Of Aisehack 2026 And Future Hackathons Under The Ai For Science Initiative.