Datasets:

Project-AgML
/

ghai_broccoli_detection

dataset_info:
  features:
    - name: image
      dtype: image
    - name: objects
      struct:
        - name: bbox
          list:
            list: float64
        - name: categories
          list:
            class_label:
              names:
                '0': Canopy
                '1': Crown
                '2': crop
  splits:
    - name: train
      num_bytes: 385632241
      num_examples: 500
  download_size: 385639627
  dataset_size: 385632241
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Ghai Broccoli Detection

A dataset for object detection of broccoli. The dataset contains 500 images with 13,303 bounding box annotations across 2 categories.

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

Citation

https://github.com/AxisAg/GHAIDatasets/blob/main/datasets/broccoli.md