Agricultural Computer Vision Dataset Survey

A catalogue of high-quality RGB image datasets of natural field scenes

Nico Heider, Lorenz Gunreben, Sebastian Zürner and Martin Schieck

Smart Farming Lab, Leipzig University

Agricultural plants — *Example images from included datasets. From left to right: LucasVision [Yo23], GlobalWheatHeadDetection [Da21], PaddyDoctor [Pe23], and CottonWeedID15 [Ch23].*

GitHub Download (.tsv) Paper arXiv

Jump to Dataset Table Jump to Survey Table

Overview

We provide access information to 45 carefully selected datasets that meet the following criteria:

Domain coherence: Natural field scenes (plants on fields or pastures taken under natural light)
High-quality ground truth data with substantial annotations
Consistent image quality (resolution, minimal motion blur, adequate lighting)
Original datasets (no web-scraped or reused images)

The datasets cover various agricultural computer vision tasks:

Weed detection and classification
Disease and pest detection
Seedling and crop detection
Plant growth stage detection
Phenotyping
Various detection and counting tasks

Citation

If you use this collection in your research, please cite our paper:


  @article{heider2025survey,
  title={A Survey of Datasets for Computer Vision in Agriculture: A catalogue of high-quality RGB image datasets of natural field scenes},
  author={Heider, Nico and Gunreben, Lorenz and Z{\"u}rner, Sebastian and Schieck, Martin},
  booktitle={45. GIL-Jahrestagung, Digitale Infrastrukturen für eine nachhaltige Land-, Forst und
Ernährungswirtschaft},
  pages={35--47},
  year={2025},
  organization={Gesellschaft f{\"u}r Informatik eV}
}

Contribute a Dataset

Are you a researcher with a high-quality agricultural computer vision dataset? We welcome contributions to our catalogue!

If you would like to have your dataset considered for inclusion in this collection, please contact us via:

E-mail
GitHub: Open an Issue

Dataset Collection

Label ↕	Year ↕	Author ↕	Plant (English) ↕	Plant (Latin) ↕	Plant Part ↕	Task ↕	Annotation ↕	Annotated Images ↕	Paper URL ↕	Dataset URL ↕

Other Surveys

Title ↕	Year ↕	Author ↕	Paper URL ↕

Acknowledgements

This work and the Rubin Feldschwarm® ÖkoSystem project are funded by the German Federal Ministry of Education and Research (BMBF) (grant no. 03RU2U051F, 03RU2U053C).