A reproducible pipeline for generating analysis-ready paddy-field segmentation datasets from cadastral vectors and Pléiades very-high-resolution imagery☆
Herman Setiadi, Muhammad Zarlis, Edy Irwansyah, Yulyani Arifin · 2026
Read A reproducible pipeline for generating analysis-ready paddy-field segmentation datasets from cadastral vectors and Pléiades very-high-resolution imagery☆ by Herman Setiadi, Muhammad Zarlis, Edy Irwansyah, Yulyani Arifin (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/PMC13451915.txt.
Description
Herman Setiadi and co-authors share in 2026 a pipeline that turns land maps and sharp satellite photos into training sets for rice field spotting. Official field border maps for five Indonesian regions were lined up with very sharp space images. The flow cuts both into paired picture and mask tiles, each about a thousand pixels square, with overlap.
Tiles with almost no rice were dropped, brightness was leveled band by band, and shape measures were logged for each field patch. The run yielded over 3,000 pairs. A small check with two vision models showed standard tools can eat the pairs at once, though the authors say the check is a smoke test, not a scoreboard. Source files sit in public stores.
Catalog record
| CREATOR | Herman Setiadi, Muhammad Zarlis, Edy Irwansyah, Yulyani Arifin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 21 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |