A Deep-Learning Framework for Land-Sliding Classification from Remote Sensing Image
Quang-Hieu Tang, Nhat-Truong Vo Dinh, Dong-Dong Pham, Quoc-Toan Nguyen, Lam Pham, Truong Nguyen · 2025
Read A Deep-Learning Framework for Land-Sliding Classification from Remote Sensing Image by Quang-Hieu Tang, Nhat-Truong Vo Dinh, Dong-Dong Pham, Quoc-Toan Nguyen, Lam Pham, Truong Nguyen (2025) 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-2507-12939.txt.
Description
Spotting landslides in satellite photos by hand is slow work after a disaster, and training data is lopsided: safe slopes vastly outnumber slides. The authors of this 2025 paper build a detection pipeline that copes with both problems. They fatten the rare class with combined online and offline image augmentation, so the model sees enough slides to learn from.
A large EfficientNet network reads each image into a compact feature signature. A support vector machine then makes the final call, a pairing meant to balance the classes and steady the verdicts against the deep net's overconfidence. On the public test set of a landslide challenge run on the Zindi platform, the system scored an F1 of about zero point eight nine. The authors frame the design as a practical recipe for rare-event mapping from orbit.
Catalog record
| CREATOR | Quang-Hieu Tang, Nhat-Truong Vo Dinh, Dong-Dong Pham, Quoc-Toan Nguyen, Lam Pham, Truong Nguyen |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 25 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-12939