Datasets:

Project-AgML
/

LeafNet_plant_disease_classification

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': diseased
            '1': healthy
    - name: species
      dtype: string
    - name: cultivar
      dtype: string
  splits:
    - name: train
      num_bytes: 7312118052
      num_examples: 4502
  download_size: 7312376792
  dataset_size: 7312118052
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 1K<n<10K

Leafnet Plant Disease Classification

This dataset comprises real-world RGB images of plant leaves, capturing both diseased and healthy specimens for disease detection tasks. Collected under typical agricultural imaging conditions, it provides a practical resource for developing computer vision models in plant pathology applications. The dataset contains 4,502 images across 2 classes: diseased, healthy.
Images per class:

diseased: 2,225
healthy: 2,277

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

Citation

<!-- TODO: add BibTeX citation -->

The dataset itself can be cited as:

Siddharth Singh Chouhan. (2022). A Database of Leaf Images: Practice towards Plant Conservation with Plant Pathology [Dataset]. Mendeley. https://doi.org/10.17632/HB74YNKJCN.5

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