A Novel Dataset for Flood Detection Robust to Seasonal Changes in Satellite Imagery

TL;DR

This paper presents a new satellite imagery dataset for flood detection, addressing the lack of suitable benchmarks, and evaluates current models, highlighting the need for advanced multimodal and temporal learning approaches.

Contribution

Introduces a novel, publicly available satellite flood dataset with diverse locations and evaluates existing models, revealing their limitations for this task.

Findings

Models showed modest performance on the dataset.

Temporal window size affects segmentation accuracy.

Current models need enhancement for flood detection.

Abstract

This study introduces a novel dataset for segmenting flooded areas in satellite images. After reviewing 77 existing benchmarks utilizing satellite imagery, we identified a shortage of suitable datasets for this specific task. To fill this gap, we collected satellite imagery of the 2019 Midwestern USA floods from Planet Explorer by Planet Labs (Image \c{opyright} 2024 Planet Labs PBC). The dataset consists of 10 satellite images per location, each containing both flooded and non-flooded areas. We selected ten locations from each of the five states: Iowa, Kansas, Montana, Nebraska, and South Dakota. The dataset ensures uniform resolution and resizing during data processing. For evaluating semantic segmentation performance, we tested state-of-the-art models in computer vision and remote sensing on our dataset. Additionally, we conducted an ablation study varying window sizes to capture temporal characteristics. Overall, the models demonstrated modest results, suggesting a requirement for future multimodal and temporal learning strategies. The dataset will be publicly available on <https://github.com/youngsunjang/SDSU_MidWest_Flood_2019>.