Datasets:

Project-AgML
/

MedLeafX_disease_classification

configs:
  - config_name: raw
    default: true
    data_dir: raw
  - config_name: augmented
    data_dir: augmented
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 10K<n<100K
dataset_info:
  - config_name: augmented
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Bacterial Spot
              '1': Healthy Leaf
              '2': Powdery Mildew
              '3': Shot Hole
              '4': Shot Hole Leaf
              '5': Yellow Leaf
  - config_name: raw
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Bacterial Spot
              '1': Healthy Leaf
              '2': Powdery Mildew
              '3': Shot Hole
              '4': Shot Hole Leaf
              '5': Yellow Leaf
      - name: species
        dtype: string

MedLeafX Disease Classification

A dataset for disease classification of 4 medicinal plant species: Camphor, Haritaki, Sojina, and Neem. The dataset contains 10,858 images across 6 classes: Bacterial Spot, Healthy Leaf, Powdery Mildew, Shot Hole, Shot Hole Leaf, Yellow Leaf.
Images per class:

Bacterial Spot: 2,408
Healthy Leaf: 3,497
Powdery Mildew: 854
Shot Hole: 1,597
Shot Hole Leaf: 834
Yellow Leaf: 1,668

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

Citation

@article{ferdous2025ai,
  title={AI-MedLeafX: a large-scale computer vision dataset for medicinal plant diagnosis},
  author={Ferdous, Md Fahim and Nissan, Faysal Bin Khaled and Nibir, Nur Muhammad and Bijoy, Md Hasan Imam},
  journal={Data in Brief},
  pages={111945},
  year={2025},
  publisher={Elsevier}
}

Ferdous, Md. Fahim; Nissan, Faysal Bin Khaled ; Nibir, Nur Muhammad ; Bijoy, Md Hasan Imam (2025), “AI-MedLeafX: A Large-Scale Computer Vision Dataset for Medicinal Plant Diagnosis”, Mendeley Data, V1, doi: 10.17632/zz7r5y4dc6.1