The dataset is currently empty. Upload or create new data files. Then, you will be able to explore them in the Dataset Viewer.

Betel Leaf Disease Dataset

Image classification dataset of betel (Piper betle) leaves for detecting common diseases, plus an unknown class of non-betel images used to make classifiers reject out-of-distribution inputs.

Classes & counts

class images
bacterial_leaf_blight 3,958
healthy_leaf 3,826
leaf_brown_spot 3,752
leaf_rot 2,538
unknown 4,040

Total: 18,114 images

Structure

Standard 🤗 ImageFolder layout — one sub-folder per class under data/.

data/
  bacterial_leaf_blight/
  healthy_leaf/
  leaf_brown_spot/
  leaf_rot/
  unknown/

Usage

from datasets import load_dataset

ds = load_dataset("KaveenDeshapriya/betel-leaf-disease")
print(ds)
print(ds["train"].features["label"].names)

Notes

  • The unknown class contains assorted non-leaf images (vehicles, animals, flowers, etc.) used as negative / out-of-distribution samples.
  • Recommended split: 70% train / 15% validation / 15% test (seed 42).

License

Released under cc-by-4.0.

Downloads last month
8