README.md

metadata

configs:
  - config_name: raw
    default: true
    data_dir: raw
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Black Spot
              '1': Healthy Leaf
              '2': Insect  Hole
              '3': Yellow Mosaic Virus
  - config_name: preprocessed
    data_dir: preprocessed
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Black Spot
              '1': Healthy Leaf
              '2': Insect Hole
              '3': Yellow Mosaic Virus
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 1K<n<10K

Rose Leaf Disease Classification

A dataset for classification of Rose leaf disease. The dataset contains raw and preprocessed versions.
The dataset contains 3,228 images.
Images per class:

• Black Spot: 409
• Healthy Leaf: 1,686
• Insect Hole: 453
• Yellow Mosaic Virus: 680

The preprocessed version has had images scaled down to 3000x3000, and backgrounds standardized.

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

Citation

@article{shacha2025roseleafinsight,
  title={RoseLeafInsight: A high-resolution image dataset for rose leaf disease recognition},
  author={Shacha, Arnob Das and Durjoy, Sabbir Hossain and ShikderKamal, Md Emon Mostafa and Shoib, Md Mehedi Hasan and Bijoy, Md Hasan Imam},
  journal={Data in Brief},
  pages={111968},
  year={2025},
  publisher={Elsevier}
}```

Shacha, Arnob Das; Durjoy, Sabbir Hossain; Shikder, Md Emon; Kamal, MD Mostafa; Shoib, Md Mehedi Hasan; Bijoy, Md Hasan Imam (2025), “RoseLeafInsight: A High-Resolution Image Dataset for Rose Leaf Disease Recognition”, Mendeley Data, V1, doi: 10.17632/8chrjdxn79.1