Recognizing protected and anthropogenic patterns in landscapes using interpretable machine learning and satellite imagery
Timo T. Stomberg, Johannes Leonhardt, Immanuel Weber, Ribana Roscher · 2023
Read Recognizing protected and anthropogenic patterns in landscapes using interpretable machine learning and satellite imagery by Timo T. Stomberg, Johannes Leonhardt, Immanuel Weber, Ribana Roscher (2023) 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/pmc-10725256.txt.
Description
Timo Stomberg and three co-authors map human impact on landscapes from space in a 2023 paper. Land cover maps guide work on biodiversity, habitat health, and climate tracking. One hard class to map is naturalness, since it lacks a crisp technical definition for a learning system to follow.
The team trains a neural network with weak labels on satellite images. Protected areas stand in for natural land, and built-up areas stand in for human influence. Tools that highlight which pixels drove each decision make the model open to inspection. A two-part network design keeps fine detail while adding a simple task layer. The result is a map that others can check and trust.
Catalog record
| CREATOR | Timo T. Stomberg, Johannes Leonhardt, Immanuel Weber, Ribana Roscher |
|---|---|
| YEAR | 2023 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 50 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |