Datasets:

Project-AgML
/

lentil_disease_classification

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
dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': Ascochyta blight
            '1': Lentil Rust
            '2': Normal
            '3': Powdery Mildew
  splits:
    - name: train
      num_bytes: 72069344
      num_examples: 5349
  download_size: 73836213
  dataset_size: 72069344

Lentil Disease Classification

A dataset for disease classification of lentil leaves. The dataset contains 5,349 images across 4 classes: Ascochyta blight, Lentil Rust, Normal, Powdery Mildew.
Images per class:

Ascochyta blight: 1,229
Lentil Rust: 1,248
Normal: 1,569
Powdery Mildew: 1,303

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

Citation

@article{mahamud2025lentil,
  title={Lentil plant disease and quality assessment: A detailed dataset of high-resolution images for deep learning research},
  author={Mahamud, Eram and Assaduzzaman, Md and Sharmin, Shayla},
  journal={Data in Brief},
  volume={58},
  pages={111224},
  year={2025},
  publisher={Elsevier}
}

Mahamud, Eram ; Tapos, Md Assaduzzaman (2024), “Lentil Plant Disease Image Dataset (4 Class)”, Mendeley Data, V2, doi: 10.17632/7vb77bz2st.2