---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: mask
    dtype: image
  - name: objects
    struct:
    - name: bbox
      list:
        list: int64
    - name: categories
      list:
        class_label:
          names:
            '0': ''
            '1': large
            '2': medium
            '3': small
  splits:
  - name: train
    num_bytes: 686518729
    num_examples: 89
  download_size: 686516546
  dataset_size: 686518729
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-4.0
task_categories:
- object-detection
size_categories:
- n<1K
---
# Avoair Db Annotated

This dataset provides real RGB imagery of avocado orchards captured from a DJI Phantom Pro 4 UAV in field conditions across the Kenitra/Allal Tazi region, Morocco. The images are annotated for object detection, focusing on avocado fruit size variations within agricultural monitoring applications. The dataset contains 89 images with 7,428 bounding box annotations across 3 categories.

This dataset is indexed on https://project-agml.github.io/ as part of the AgML python library.

## Citation

```bibtex
@article{elamraoui2022avo,
  title={Avo-AirDB: An avocado UAV Database for agricultural image segmentation and classification},
  author={EL Amraoui, Khalid and Lghoul, Mouataz and Ezzaki, Ayoub and Masmoudi, Lhoussaine and Hadri, Majid and Elbelrhiti, Hicham and Simo, Aziz Abdou},
  journal={Data in Brief},
  volume={45},
  pages={108738},
  year={2022},
  publisher={Elsevier}
}
```


*This dataset was reformatted from its original format to match HuggingFace standards.*