# Deglet Nour Date Palm Bunch Detection: 5,530 Field Images with 24,559 Bounding Boxes from Kebili Oases, Tunisia (AgML Mirror)

## Facts

- Provider: Ahlem Maghzaoui, Emna Aridhi, Sadok Ben Yahia, Sahbi Bahroun and Abdelkader Mami (Faculty of Sciences of Tunis); Hugging Face mirror by Project-AgML
- License: CC-BY-4.0 (https://zenodo.org/records/15387868)
- Size: 4 files
- Format: Parquet (in sampled files)
- Stored at: Hugging Face, https://huggingface.co/datasets/Project-AgML/date_cluster_detection
- Access: public, readable without an account
- DOI: https://doi.org/10.5281/zenodo.15387868
- Category: Agriculture & Environment
- Source: https://huggingface.co/datasets/Project-AgML/date_cluster_detection
- Page: https://scigantic.com/archive/345d5c82-2b9d-515b-9d61-fde7587876d7
- Keywords: Agriculture & Environment, image segmentation, benchmark, Parquet, Hugging Face

## About

High-resolution images of Deglet Nour date palm (Phoenix dactylifera) fruit bunches collected in oases of the Kebili governorate, southern Tunisia, during the 2023 harvest season, under real field conditions with variable lighting, occlusion and bunch covering. Built as a benchmark for AI-guided smart and autonomous date harvesting. This mirror holds 5,530 images with 24,559 bunch bounding boxes in one class. Useful for date bunch detection, harvesting robot vision, keypoint and segmentation follow-on work, and yield estimation in date palm orchards. Stored as Parquet shards with the image bytes and annotations in columns; load with datasets.load_dataset("Project-AgML/date_cluster_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 Ahlem Maghzaoui, Emna Aridhi, Sadok Ben Yahia, Sahbi Bahroun and Abdelkader Mami et al. and the DOI 10.5281/zenodo.15387868.

## Load the data

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

## Data structure

File types in sampled files: Parquet

Example file: `data/train-00000-of-00002.parquet`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/345d5c82-2b9d-515b-9d61-fde7587876d7
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/345d5c82-2b9d-515b-9d61-fde7587876d7
- Over MCP: `get_schema_card` with id `345d5c82-2b9d-515b-9d61-fde7587876d7` at https://api.scigantic.com/mcp

## Related topics

- [Parquet datasets](https://scigantic.com/topics/parquet)
- [Benchmark datasets](https://scigantic.com/topics/benchmark)
- [Image segmentation datasets](https://scigantic.com/topics/image-segmentation)
- [Agriculture & Environment datasets](https://scigantic.com/topics/agriculture-and-environment)
