The dataset can be used freely for academic purposes. Please send your full name, affiliation and intended use of the dataset to Philipp Lottes or Jan Weyler. Make sure to use your institutional e-mail address. We will send you a download link to get the dataset and provide you more information about the annotation format. The dataset contains orthorectified images of breeding plots captured at five different dates. Thus, it contains sugar beets at a variety of growth stages. Below you can find an example of an input image and its annotation.
Growth Stages
Our dataset contains plants at different growth stages. In the following figure we provide an overview of the growth stages for each session of data acquisition.
Statistics
In the following we provide a statistical overview for the train, val, and test split.
Please note that there is no overlap between the splits.
| Train | Val | Test | |
| # Total sugar beets | 1673 | 840 | 769 |
| # Total sugar beet leaves | 6584 | 3410 | 2929 |
| # Average leaves per beet | 3.94 | 4.06 | 3.81 |
| Std. leaves per beet | 1.27 | 1.23 | 1.18 |
| # Maximum leaves per beet | 11 | 12 | 9 |
| # Minimum leaves per beet | 1 | 1 | 1 |
Train Images
In the following you can find all 10 images used for training.
Each image has a size of 4320 [px] x 4100 [px].
Validation Images
In the following you can find all 5 images used for validation.
Each image has a size of 4320 [px] x 4100 [px].
Test Images
In the following you can find all 5 test images.
Each image has a size of 4320 [px] x 4100 [px].
License
All datasets on this page are copyright by Rheinische Friedrich-Wilhelms-Universität Bonn and published under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License. This means that you must attribute the work in the manner specified by the authors, you may not use this work for commercial purposes and if you alter, transform, or build upon this work, you may distribute the resulting work only under the same license. Per GDPR requirements, to download and use the data you need to register and specify the intended purpose of using the dataset.