From Image Series to Resilient Varieties: Transformer‐Based Canopy Cover Extraction and Weather‐Informed Nonlinear Modeling
Summary statement Green canopy cover dynamics extracted from high‐throughput RGB imagery via automated labeled segmentation enabled nonlinear modeling that accounted for short‐term weather variation, explaining up to 56% of protein yield variation in 150 soybean genotypes across 10 seasons and enabling prediction in new environments as well as identification of tolerant genotypes.
Authors
- Paul Stephan (ORCID: https://orcid.org/0000-0002-9197-7567)
- Andreas Hund (ORCID: https://orcid.org/0000-0002-2309-1625)
- Fabio Sigrist (ORCID: https://orcid.org/0000-0002-3994-2244)
- Christoph Barendregt
- Martin Mächler (ORCID: https://orcid.org/0000-0002-8685-9910)
- Achim Walter (ORCID: https://orcid.org/0000-0001-7753-9643)
- Beat Keller (ORCID: https://orcid.org/0000-0001-5852-6660)
- Corina Oppliger (ORCID: https://orcid.org/0000-0002-2866-6061)
- Claude‐Alain Bétrix
- M. Lob
- Tatiana Bielakova
- Jovin Koller
Institutions
- ETH Zurich (CH)
- Institute for Biomedical Engineering (CH)
- Zurich University of Teacher Education (CH)
- Agroscope (CH)
- Delley Seeds and Plants (Switzerland) (CH)
Publication Details
- Journal
- Plant Cell & Environment
- Published
- 2026-10-12
- DOI
- https://doi.org/10.1111/pce.70947
- Primary Topic
- Remote Sensing in Agriculture
- Type
- article
- Field-Weighted Citation Impact
- 0.00