Project-AgML/BDHerbalPlants_variety_classification

BDHerbalPlants Variety Classification

A dataset for variety classification of herbal plant leaves. The dataset contains raw and augmented versions.
The raw dataset contains 1,792 images.
Images per class:

• Datura (Datura stramonium): 238
• Dhonya Pata ( Coriandrum Sativum): 223
• Kalokeshi: 243
• Neem (Azadirachta indica): 214
• Pathor Kuchi (Kalanchoe pinnata): 189
• Pudina (Mentha Arvensis): 222
• Thankuni (Centella asiatica): 236
• Tulsi (Ocimum tenuiflorum): 227

The augmented dataset contains 14,336 images.
Images per class:

• Datura (Datura stramonium): 1,904
• Dhonya Pata ( Coriandrum Sativum): 1,784
• Kalokeshi (Eclipta prostrata): 1,944
• Neem (Azadirachta indica): 1,712
• Pathor Kuchi (Kalanchoe pinnata): 1,512
• Pudina (Mentha Arvensis): 1,776
• Thankuni (Centella asiatica): 1,888
• Tulsi (Ocimum tenuiflorum): 1,816

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

Citation

@article{khandaker2025bdherbalplants,
  title={BDHerbalPlants: augmented and curated herbal plants image dataset for classification.},
  author={Khandaker, Sunzil and Rahman, Md Mizanur},
  journal={Data in Brief},
  pages={111885},
  year={2025},
  publisher={Elsevier}
}

Khandaker, Sunzil; Rahman, Md Mizanur (2025), “BDHerbalPlants”, Mendeley Data, V3, doi: 10.17632/md59kt54jy.3