---
license: cc-by-4.0
language:
- en
tags:
- soybean
- segmentation
- canopy_cover
- breeding
- phenotyping
- time-series
size_categories:
- 1K<n<10K
---
raw             |  segmentation | plot canopy cover
:-------------------------:|:-------------------------:|:-------------------------:
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6304a1226dbbb80f16365385/uruuOAomtl3qLDjuiHSwA.jpeg)|![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6304a1226dbbb80f16365385/blDSXNUHH6hJpLUElLr42.jpeg)|![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/6304a1226dbbb80f16365385/_fyFqEUtVQK-ofW_jHGzG.jpeg)



We introduce a longitudinal field dataset to analyze soybean growth under varying environmental conditions. 
It includes 17,247 high-resolution top-of-canopy RGB images collected across eight growing seasons in Eschikon, Switzerland, complemented by yield data and hourly weather data.
The dataset captures temporal variation in genotype responses to environmental stress, providing insights into growth dynamics and genotype by environment interactions. 
High spatio-temporal resolution supports the identification of stress-tolerant genotypes and contributes to improved yield prediction and stability.

The data, including the images and segmentation masks, are available at the [ETH research collection](https://doi.org/10.3929/ethz-b-000742401).  
Either copy the data to `/data` or replace `/data` in the file paths to your data folder directoy using a map call.


A more detailed readme will be added once the paper pre-print has been released.