FLNet: Flood-Induced Agriculture Damage Assessment using Super Resolution of Satellite Images

2026-01-07

arXiv.org

AI Summary

Traditional post-flood crop damage assessment via manual surveys is inefficient and prone to subjective bias, while existing satellite-based methods often suffer from cloud cover and low spatial resolution, hindering accurate and rapid evaluation. To address these limitations, this work proposes FLNet—a deep learning framework that jointly optimizes super-resolution and damage classification. By leveraging open-access Sentinel-2 imagery (10 m resolution), FLNet first reconstructs images to 3 m resolution through super-resolution and then performs damage classification in an end-to-end trainable pipeline. Evaluated on the BFCD-22 dataset, the proposed method improves the F1-score for the “complete damage” class from 0.83 to 0.89, achieving accuracy comparable to commercial high-resolution imagery. This approach establishes a new, cost-effective paradigm for nationwide automated agricultural disaster assessment with high precision.

Technology Category

• Computer Vision: Learning & Optimization for CV
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Distributing government relief efforts after a flood is challenging. In India, the crops are widely affected by floods; therefore, making rapid and accurate crop damage assessment is crucial for effective post-disaster agricultural management. Traditional manual surveys are slow and biased, while current satellite-based methods face challenges like cloud cover and low spatial resolution. Therefore, to bridge this gap, this paper introduced FLNet, a novel deep learning based architecture that used super-resolution to enhance the 10 m spatial resolution of Sentinel-2 satellite images into 3 m resolution before classifying damage. We tested our model on the Bihar Flood Impacted Croplands Dataset (BFCD-22), and the results showed an improved critical"Full Damage"F1-score from 0.83 to 0.89, nearly matching the 0.89 score of commercial high-resolution imagery. This work presented a cost-effective and scalable solution, paving the way for a nationwide shift from manual to automated, high-fidelity damage assessment.

Problem

• flood damage assessment
• agricultural damage
• satellite imagery
• spatial resolution
• crop monitoring

Innovation

• super-resolution
• deep learning
• flood damage assessment
• Sentinel-2
• agricultural remote sensing