Croppie Coffee Cherry Detection Dataset: Smartphone Photos of Arabica and Robusta Branches with Ripeness-Class Bounding Boxes, Uganda (Producers Direct and Alliance Bioversity-CIAT)

Provider
Producers Direct and Alliance of Bioversity International and CIAT
License
CC-BY-SA-4.0 (source)
Size
13 files
Stored at
Hugging Face (https://huggingface.co/datasets/rgautroncgiar/croppie_coffee_ug)
Access
Free and public. No account with the data provider is needed.
Category
Agriculture & Environment
Source
https://huggingface.co/datasets/rgautroncgiar/croppie_coffee_ug

About this dataset

On-farm smartphone photographs of Arabica and Robusta coffee tree branches in Uganda with a bounding box around every coffee cherry, built for Croppie, a machine vision tool for coffee yield estimation by Producers Direct and the Alliance of Bioversity International and CIAT, funded through GIZ's Fair Forward initiative. The original 633 images carry about 61,050 cherry boxes; cherries over 10 mm are labeled by ripeness color as green, yellow, red or dark brown (overripe or dry), with an extra class for low-visibility or uncertain cherries. Images were cut into 480 x 640 tiles plus a downscaled full view, giving 5,836 training and 2,497 validation images in YOLO format. The repository (about 0.9 GB) holds data/train.zip and data/val.zip, classes.json mapping class ids, the annotation guide and a visualization script. Useful for fruit detection and counting, ripeness classification and smallholder crop yield forecasting with object detectors such as YOLO. Croppie data © 2024 Producers Direct and Alliance Bioversity & CIAT, licensed CC BY-SA 4.0.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="rgautroncgiar/croppie_coffee_ug", filename="data/train.zip", repo_type="dataset")
print(path)

Data structure

Files

Example file: data/train.zip

Analyze this dataset without downloading it

On Scigantic, Croppie Coffee Cherry Detection Dataset: Smartphone Photos of Arabica and Robusta Branches with Ripeness-Class Bounding Boxes, Uganda (Producers Direct and Alliance Bioversity-CIAT) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Agriculture & Environment datasets