# Seed Germination Detection Time Series: 23,797 Petri Dish Images of Maize, Rye and Pearl Millet Seeds with Germinated Bounding Boxes (TU Munich, AgML Mirror)

## Facts

- Provider: Nikita Genze and Dominik Grimm (Technical University of Munich, Campus Straubing); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://data.mendeley.com/public-api/datasets/4wkt6thgp6?version=3)
- Size: 6 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/seed_germination_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.17632/4wkt6thgp6.3
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/seed_germination_detection
- Page: https://scigantic.com/archive/ccdf0990-9e0d-52f6-a86a-822d4e606f2a
- Keywords: Agriculture & Environment, time series, Parquet, Hugging Face

## About

Time-lapse images from high-throughput seed germination experiments on 824 maize (Zea mays), 811 rye (Secale cereale) and 814 pearl millet (Pennisetum glaucum) seeds. About 10 seeds per petri dish were laid on black cloth and photographed every 30 minutes for about two days with a low-cost Raspberry Pi Camera Module v2.1. Every seed is boxed and labeled germinated (radicle emerged through the seed coat) or not germinated, giving 23,797 images (624 x 624, one dish each) with 235,933 boxes. Crop, dish series, time step and the authors' train, validation and test split (by dish series) are kept as columns. The data support machine-learning germination counting, germination-time prediction and seed quality assessment in place of manual scoring. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/seed_germination_detection") or through the AgML Python library (Project-AgML, UC Davis), which reformatted the original release. License: CC BY 4.0, as stated by the original publisher; cite Nikita Genze and Dominik Grimm et al. and the DOI 10.17632/4wkt6thgp6.3.

## Load the data

```python
from datasets import load_dataset
ds = load_dataset("Project-AgML/seed_germination_detection")
print(ds)
```

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00004.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/ccdf0990-9e0d-52f6-a86a-822d4e606f2a
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/ccdf0990-9e0d-52f6-a86a-822d4e606f2a
- Over MCP: `get_schema_card` with id `ccdf0990-9e0d-52f6-a86a-822d4e606f2a` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Time series datasets](https://scigantic.com/topics/time-series)
- [Agriculture & Environment datasets](https://scigantic.com/topics/agriculture-and-environment)
