Flood Data Analysis on SpaceNet 8 Using Apache Sedona

TL;DR

The paper addresses flood detection in SpaceNet8 and the challenge of model errors by leveraging Apache Sedona for scalable geospatial data processing. It introduces a framework that retrieves historical flood cases, adapts them to current scenarios, and uses clustering to guide model revisions, enabling targeted error analysis. Empirical results show improvements in precision, F1, and IoU after error-driven refinements, along with insights into data quality issues and the impact of contrast enhancement. This approach enhances explainability and robustness for satellite-based flood monitoring with practical implications for disaster response and infrastructure resilience.

Abstract

With the escalating frequency of floods posing persistent threats to human life and property, satellite remote sensing has emerged as an indispensable tool for monitoring flood hazards. SpaceNet8 offers a unique opportunity to leverage cutting-edge artificial intelligence technologies to assess these hazards. A significant contribution of this research is its application of Apache Sedona, an advanced platform specifically designed for the efficient and distributed processing of large-scale geospatial data. This platform aims to enhance the efficiency of error analysis, a critical aspect of improving flood damage detection accuracy. Based on Apache Sedona, we introduce a novel approach that addresses the challenges associated with inaccuracies in flood damage detection. This approach involves the retrieval of cases from historical flood events, the adaptation of these cases to current scenarios, and the revision of the model based on clustering algorithms to refine its performance. Through the replication of both the SpaceNet8 baseline and its top-performing models, we embark on a comprehensive error analysis. This analysis reveals several main sources of inaccuracies. To address these issues, we employ data visual interpretation and histogram equalization techniques, resulting in significant improvements in model metrics. After these enhancements, our indicators show a notable improvement, with precision up by 5%, F1 score by 2.6%, and IoU by 4.5%. This work highlights the importance of advanced geospatial data processing tools, such as Apache Sedona. By improving the accuracy and efficiency of flood detection, this research contributes to safeguarding public safety and strengthening infrastructure resilience in flood-prone areas, making it a valuable addition to the field of remote sensing and disaster management.

Figures (7)

• Figure 1: Overview of Our Work
• Figure 2: The model identified four primary error types: (a)(b) Target omission due to small object size; (c)(d) Spatial confusion, where targets blend with surroundings; (e)(f) Missing information from incomplete satellite imagery labels; (g)(h) Incorrect labels from manual annotation errors. The illustration shows target objects (red areas), actual labels ("True label"), and model predictions ("Pred label").
• Figure 3: Comparison between satellite images, annotated images, and Baseline prediction images. From top to bottom are examples of Cluster 0, Cluster 1, and Cluster 2.
• Figure 4: Clustering results of the Baseline model and the SOTA model. (a) Baseline model, (b) SOTA model.
• Figure 5: Comparison between satellite images, annotated images, and SOTA prediction images. From top to bottom are examples of Cluster 0, Cluster 1.