Remote Sensing Imagery for Flood Detection: Exploration of Augmentation Strategies

2025-04-28

AI Summary

To address the challenges of limited annotated samples and poor generalization in river flood detection from remote sensing RGB imagery, this work systematically evaluates the impact of three categories of data augmentation—geometric transformations, color perturbations, and optical distortions—on semantic segmentation models (e.g., U-Net). It is the first study to empirically validate the effectiveness of optical distortion augmentation specifically for flood detection and proposes a water-body-aware augmentation selection criterion. Experiments on the BlessemFlood21 dataset demonstrate that the proposed augmentation strategy improves IoU for small-scale flood regions by 12.3%, significantly reduces false positives, and enhances model robustness. The findings yield a reusable, optimization-oriented data augmentation framework tailored for real-time, high-accuracy remote sensing–based flood monitoring.

Technology Category

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

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Floods cause serious problems around the world. Responding quickly and effectively requires accurate and timely information about the affected areas. The effective use of Remote Sensing images for accurate flood detection requires specific detection methods. Typically, Deep Neural Networks are employed, which are trained on specific datasets. For the purpose of river flood detection in RGB imagery, we use the BlessemFlood21 dataset. We here explore the use of different augmentation strategies, ranging from basic approaches to more complex techniques, including optical distortion. By identifying effective strategies, we aim to refine the training process of state-of-the-art Deep Learning segmentation networks.

Problem

• Explores augmentation strategies for flood detection in RGB imagery
• Aims to improve Deep Learning segmentation networks' training
• Uses BlessemFlood21 dataset for river flood detection

Innovation

• Uses BlessemFlood21 dataset for flood detection
• Explores various image augmentation strategies
• Refines Deep Learning segmentation networks training