# Banana Bunch Detection Images from Four Fields in Madeira, Portugal: 2,179 Smartphone Photos with 3,718 Boxes (Universidade da Madeira, AgML Mirror)

## Facts

- Provider: Preety Baglat, Ahatsham Hayat, Sheikh Shanawaz Mostafa, Fábio Mendonça and Fernando Morgado Dias (Universidade da Madeira; ITI/LARSyS); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://zenodo.org/records/15642838)
- Size: 42 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/banana_bunch_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.15642838
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/banana_bunch_detection
- Page: https://scigantic.com/archive/026a5469-d328-5423-b489-e8f4345f3745
- Keywords: Agriculture & Environment, Parquet, Hugging Face

## About

Images of banana bunches on plants in four fields on the island of Madeira, Portugal (Câmara de Lobos, Lugar de Baixo, Ponta do Sol and Santo António), taken on several dates between May and October 2022 with different smartphones (Samsung Galaxy A12 and Note 9, OnePlus 9, iPhone 12) under varied lighting and weather. Bunches were labeled manually; the mirror holds 2,179 images with 3,718 bounding boxes in one class. Useful for banana bunch detection, plantation monitoring, harvest planning and benchmarking detectors across fields and cameras. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/banana_bunch_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 Preety Baglat, Ahatsham Hayat, Sheikh Shanawaz Mostafa, Fábio Mendonça and Fernando Morgado Dias et al. and the DOI 10.5281/zenodo.15642838.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00040.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/026a5469-d328-5423-b489-e8f4345f3745
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/026a5469-d328-5423-b489-e8f4345f3745
- Over MCP: `get_schema_card` with id `026a5469-d328-5423-b489-e8f4345f3745` 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)
