Urban Flood Observations (UFO): A hand-labeled training and validation dataset of post-flood inundation

2026-04-24

AI Summary

This study addresses the challenge of accurately identifying urban flood inundation areas in satellite imagery, which is hindered by low spatial resolution, infrequent revisit intervals, and persistent cloud cover. To overcome these limitations, the authors present UFO—the first global, high-resolution, multi-event, and manually annotated urban flood dataset—comprising 215 three-meter PlanetScope image chips from 14 flood events between 2017 and 2021, each accompanied by binary inundation labels. Using leave-one-event-out cross-validation, segmentation models trained on UFO achieve an average Intersection over Union (IoU) of 77.3%, substantially outperforming NASA IMPACT (44.1%) and Dynamic World (48.1%). This benchmark establishes a new standard for flood monitoring and evaluation of water-related remote sensing products.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Urban flooding affects lives and infrastructure worldwide. Mapping inundation in complex urban environments from satellite imagery remains challenging due to limited spatial resolution, infrequent acquisitions, and cloud cover. We present Urban Flood Observations (UFO), a global, hand-labeled dataset of post-flood inundation in diverse urban settings. UFO comprises 215 image chips (1024 by 1024 pixels) from 14 flood events between 2017 and 2021, derived from 3 m PlanetScope imagery. Each chip is annotated with two classes: 'inundated' (all visible surface water, including floodwater and pre-existing water bodies (permanent or seasonal)) and 'non-inundated'. To demonstrate the dataset's utility, we trained a segmentation model using leave-one-event-out cross-validation, achieving a mean Intersection over Union (IoU) of 77.3. We also used UFO to evaluate two widely used surface water products, the Sentinel-1-based NASA IMPACT model and Google's 10 m Dynamic World water class, which yielded IoUs of 44.1 and 48.1, respectively. UFO is publicly available to support the development and validation of urban inundation mapping methods.

Problem

• urban flooding
• inundation mapping
• satellite imagery
• flood monitoring
• urban environments

Innovation

• urban flood mapping
• hand-labeled dataset
• high-resolution satellite imagery
• semantic segmentation
• flood inundation validation