Evaluating Vision-Language Model Detection of Flowering and Fruit Development in Seminole Pumpkin
This pilot study will test whether a vision-capable language model can detect the first open flower and first visible fruit on individually tagged Seminole pumpkin plants using fixed-camera images. Model predictions will be compared with independent field observations and human image labels, with results reported by plant, event type, and detection error. The study is designed as a small, reproducible feasibility evaluation; its sample is too limited to support broad claims about other crops or growing conditions.
Authors
- Peter McGrath
Institutions
- Appalachian State University (US)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.5281/zenodo.24320200
- Primary Topic
- Smart Agriculture and AI
- Type
- article
- Field-Weighted Citation Impact
- 0.00