FLNet: Flood-Induced Agriculture Damage Assessment using Super Resolution of Satellite Images
Sanidhya Ghosal, Anurag Sharma, Sushil Ghildiyal, Mukesh Saini · 2026
Read FLNet: Flood-Induced Agriculture Damage Assessment using Super Resolution of Satellite Images by Sanidhya Ghosal, Anurag Sharma, Sushil Ghildiyal, Mukesh Saini (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-03884.txt.
Description
Sanidhya Ghosal and three co-authors built FLNet, a deep learning system that assesses flood damage to crops from satellite images, to speed government relief in India. Manual damage surveys are slow and biased, while satellite methods battle cloud cover and coarse resolution. FLNet first uses super-resolution to sharpen Sentinel-2 images from ten meter to three meter detail, then classifies the damage in the enhanced images.
Tested on the Bihar Flood Impacted Croplands dataset, the sharpening step clearly improved detection of the critical full damage category. Its score nearly matched what commercial high-resolution imagery achieved, without the commercial price. The authors present the approach as cost-effective and scalable, a candidate for a nationwide shift from manual surveys to automated, high-fidelity assessment after floods, when relief decisions cannot wait for field teams.
Catalog record
| CREATOR | Sanidhya Ghosal, Anurag Sharma, Sushil Ghildiyal, Mukesh Saini |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 27 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2601-03884