Datasets:

Project-AgML
/

maize_disease_classification

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': Healthy
            '1': MLN
            '2': MSV
  splits:
    - name: train
      num_bytes: 7911968327
      num_examples: 9356
  download_size: 7528242288
  dataset_size: 7911968327
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 10K<n<100K

Maize Disease Classification

A dataset for disease classification of Maize leaves. The dataset contains 9,356 images across 3 classes: Healthy, MLN, MSV.
Images per class:

Healthy: 3,073
MLN: 3,231
MSV: 3,052

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

Citation

@article{mduma2024updating,
  title={Updating “machine learning imagery dataset for maize crop: A case of Tanzania” with expanded data to cover the new farming season},
  author={Mduma, Neema and Mayo, Flavia},
  journal={Data in Brief},
  volume={54},
  pages={110359},
  year={2024},
  publisher={Elsevier}
}

Mduma, Neema; Mayo, Flavia (2023), “Maize Imagery Dataset - Tanzania”, Mendeley Data, V1, doi: 10.17632/fkw49mz3xs.1