# ImageWeeds: Aerial and Greenhouse Weed Images with 6,932 Bounding Boxes for Spot Spraying, 3,208 Images (North Dakota State University, AgML Mirror)

## Facts

- Provider: Nitin Rai, Maria Villamil Mahecha, Annika Christensen, Jamison Quanbeck, Kirk Howatt and Michael Ostlie (North Dakota State University); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://data.mendeley.com/public-api/datasets/8kjcztbjz2?version=2)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/ImageWeeds_weed_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.17632/8kjcztbjz2.2
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/ImageWeeds_weed_detection
- Page: https://scigantic.com/archive/f87e6a70-dfc7-56b1-8bdf-5f440d67ba07
- Keywords: Agriculture & Environment, Parquet, Hugging Face

## About

ImageWeeds holds aerial and greenhouse images of weeds in several formats, collected at North Dakota State University to develop computer vision for real-time weed identification and site-specific (spot) herbicide spraying. This mirror holds the detection subset: 3,208 images with 6,932 bounding boxes across five weed categories. A separate, smaller aerial subset is mirrored as Project-AgML/ImageWeeds_aerial_weed_detection. Useful for weed detection under greenhouse and drone viewpoints and for spray-targeting research. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/ImageWeeds_weed_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 Nitin Rai, Maria Villamil Mahecha, Annika Christensen, Jamison Quanbeck, Kirk Howatt and Michael Ostlie et al. and the DOI 10.17632/8kjcztbjz2.2.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00002.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/f87e6a70-dfc7-56b1-8bdf-5f440d67ba07
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/f87e6a70-dfc7-56b1-8bdf-5f440d67ba07
- Over MCP: `get_schema_card` with id `f87e6a70-dfc7-56b1-8bdf-5f440d67ba07` at https://api.scigantic.com/mcp

## Related topics

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