# Horseradish and Weed Detection Images from Commercial Fields in Southern Illinois: 2,697 Images with 25,298 YOLO Boxes (UIUC, AgML Mirror)

## Facts

- Provider: Abhinav Pagadala and Sunoj Shajahan (University of Illinois Urbana-Champaign); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://data.mendeley.com/public-api/datasets/fcf7brsfm6?version=1)
- Size: 3 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/horseradish_weed_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.17632/fcf7brsfm6.1
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/horseradish_weed_detection
- Page: https://scigantic.com/archive/2c9a6c4b-ff7f-5eb5-aeb2-d73b318bf531
- Keywords: Agriculture & Environment, Parquet, Hugging Face

## About

Annotated RGB images for benchmarking YOLO weed detectors in commercial horseradish (Armoracia rusticana) production, a high-value specialty crop with few herbicide options. Images were captured in the 2024 season in two commercial fields in Collinsville, Illinois, and a research plot at the Illinois Autonomous Farm (UIUC), from a handheld iPhone 13 Mini on a monopod between rows and from a Farm-ng Amiga robot with Luxonis OAK-D cameras; video frames were extracted, curated and augmented. Two classes are labeled: horseradish (crop) and weed, a composite class covering waterhemp (Amaranthus tuberculatus), Palmer amaranth (Amaranthus palmeri) and other broadleaf and grass weeds. The mirror holds 2,697 images with 25,298 boxes. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/horseradish_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 Abhinav Pagadala and Sunoj Shajahan et al. and the DOI 10.17632/fcf7brsfm6.1.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00001.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/2c9a6c4b-ff7f-5eb5-aeb2-d73b318bf531
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/2c9a6c4b-ff7f-5eb5-aeb2-d73b318bf531
- Over MCP: `get_schema_card` with id `2c9a6c4b-ff7f-5eb5-aeb2-d73b318bf531` 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)
