Introduction
We are building a collaborative, AI-ready database of plants from around the world to support the development of Plant Doctor, a smart farming tool that quantifies plant health in real time. This database includes images, sensor readings, and environmental data covering a wide range of species, conditions, and regions, making it a valuable resource for training robust machine learning models.
Link to the article: https://doi.org/10.1016/j.measurement.2025.117094
How to collaborate?
To collaborate with us, please contact umeshin[at]waseda.jp. We will provide you with access to a secure Box folder where you can upload your data.
Currently, we are sharing our own datasets for both machine vision and image segmentation tasks. Accepted image formats include JPG, PNG, and TIFF.
For annotations, we use the PASCAL VOC format (.xml files).
Our database is being published on Zenodo, an open-access research data repository supported by CERN and the European Commission. Zenodo ensures long-term preservation, provides DOIs for citation, and supports FAIR data principles (Findable, Accessible, Interoperable, Reusable), making it an ideal platform for sharing scientific datasets with the global research community.
Databases
Currently available databases published for your use: