Datasets:

Project-AgML
/

ERWIAM_blight_detection

dataset_info:
  features:
    - name: image
      dtype: image
    - name: objects
      struct:
        - name: bbox
          list:
            list: float64
        - name: categories
          list:
            class_label:
              names:
                '0': Flower
                '1': Shoot
                '2': Maybe
                '3': Leaf
  splits:
    - name: train
      num_bytes: 360647800
      num_examples: 1698
  download_size: 392683479
  dataset_size: 360647800
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - object-detection
size_categories:
  - 1K<n<10K

Erwiam Blight Detection

A dataset for detection of fire blight in an apple orchard. The dataset contains 1,698 images with 15,761 bounding box annotations across 4 categories.

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

Citation

@article{mass2024annotated,
  title={Annotated image dataset of fire blight symptoms for object detection in orchards},
  author={Ma{\ss}, Virginia and Alirezazadeh, Pendar and Seidl-Schulz, Johannes and Leipnitz, Matthias and Fritzsche, Eric and Ibraheem, Rasheed Ali Adam and Geyer, Martin and Pflanz, Michael and Reim, Stefanie},
  journal={Data in Brief},
  volume={56},
  pages={110826},
  year={2024},
  publisher={Elsevier}
}

Maß, Virginia; Alirezazadeh, Pendar; Seidl-Schulz, Johannes; Leipnitz, Matthias; Fritzsche, Eric; Ibraheem, Rasheed Ali Adam; Geyer, Martin; Pflanz, Michael; Reim, Stefanie (2024), “ERWIAM dataset”, Mendeley Data, V1, doi: 10.17632/fpmnncmg84.1

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