# Weed and Crop Detection in Precision Agriculture: 1,120 RGB Images with 17,693 YOLO Boxes for Five Weeds and Eight Crops (North Dakota State University, AgML Mirror)

## Facts

- Provider: Arjun Upadhyay, Sunil G C, Maria Villamil Mahecha, Joseph Mettler, Kirk Howatt, William Aderholdt, Michael Ostlie and Xin Sun (North Dakota State University); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://data.mendeley.com/public-api/datasets/mthv4ppwyw?version=2)
- Size: 18 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/weed_crop_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.17632/mthv4ppwyw.2
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/weed_crop_detection
- Page: https://scigantic.com/archive/89d1f68f-ea01-550b-b22d-5d2c9a6d89bc
- Keywords: Agriculture & Environment, Parquet, Hugging Face

## About

RGB images of five weed species (kochia, horseweed, waterhemp, common ragweed and redroot pigweed) and eight crops (black bean, canola, corn, lentil, field pea, flax, soybean and sugar beet), labeled with bounding boxes for 13 classes, collected at North Dakota State University with USDA and NIFA funding to train detectors for robotic weed control and site-specific herbicide spraying. The mirror holds 1,120 images with 17,693 boxes; the original release gives the labels in YOLO format. Useful for crop and weed discrimination, multi-species weed identification, and benchmarking object detectors for Northern Great Plains row and broadleaf crops. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/weed_crop_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 Arjun Upadhyay, Sunil G C, Maria Villamil Mahecha, Joseph Mettler, Kirk Howatt, William Aderholdt, Michael Ostlie and Xin Sun et al. and the DOI 10.17632/mthv4ppwyw.2.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00016.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/89d1f68f-ea01-550b-b22d-5d2c9a6d89bc
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/89d1f68f-ea01-550b-b22d-5d2c9a6d89bc
- Over MCP: `get_schema_card` with id `89d1f68f-ea01-550b-b22d-5d2c9a6d89bc` 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)
