# Grape Bunch and Vine Trunk Object Detection Images: 2,846 RGB Vineyard Photos from the Douro Region, Portugal (INESC TEC, AgML Mirror)

## Facts

- Provider: André Silva Aguiar and Sandro Magalhães (INESC TEC); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://zenodo.org/records/5139598)
- Size: 12 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/vineyard_grape_trunk_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.5139598
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/vineyard_grape_trunk_detection
- Page: https://scigantic.com/archive/ef672dd5-bc79-5e97-bfd2-ac2eba379711
- Keywords: Agriculture & Environment, Parquet, Hugging Face

## About

Ground-level RGB photographs of vineyards in the Douro wine region of Portugal, taken with several cameras under varied illumination, with bounding boxes for three classes: vine trunks, small grape bunches just after bloom, and medium-sized grape bunches. The release accompanies work on detecting grape clusters with quantized deep-learning detectors on low-power hardware for agricultural robots. The raw configuration holds 2,846 unaugmented images with 13,218 boxes; the augmented configuration holds 18,515 copies (rotation, flipping, hue and saturation shifts, noise, scaling and translation) with 85,722 boxes, each labeled with its augmentation and capture set. Thermal images in the original Zenodo record are not included. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/vineyard_grape_trunk_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 André Silva Aguiar and Sandro Magalhães et al. and the DOI 10.5281/zenodo.5139598.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `augmented/train-00000-of-00006.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/ef672dd5-bc79-5e97-bfd2-ac2eba379711
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/ef672dd5-bc79-5e97-bfd2-ac2eba379711
- Over MCP: `get_schema_card` with id `ef672dd5-bc79-5e97-bfd2-ac2eba379711` 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)
