---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': bacterial_leaf_blight
          '1': brown_spot
          '2': healthy
          '3': leaf_blast
          '4': leaf_scald
          '5': narrow_brown_spot
  splits:
  - name: train
    num_bytes: 90185777
    num_examples: 773
  download_size: 90205016
  dataset_size: 90185777
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-4.0
task_categories:
- image-classification
size_categories:
- n<1K
---
# Rice Leaf Disease Classification Bd

This dataset features real RGB images of rice leaves collected in field environments across Bangladesh during the Aman 2025 rice season. Captured using a handheld iPhone 12 Pro, the images depict various disease conditions on rice leaves under natural agricultural settings. The collection provides a practical resource for developing and evaluating computer vision models focused on rice disease detection. The dataset contains 773 images across 6 classes: bacterial_leaf_blight, brown_spot, healthy, leaf_blast, leaf_scald, narrow_brown_spot.  
Images per class:
- bacterial_leaf_blight: 116
- brown_spot: 190
- healthy: 110
- leaf_blast: 133
- leaf_scald: 107
- narrow_brown_spot: 117

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

## Citation

```bibtex
@article{masud2026iot,
  title={IoT-RiceMobileNet: An improved lightweight MobileNetV2 Model for real-time multi-class rice disease detection using IoT},
  author={Masud, Khawja Imran and Zihad, Md Yasin and Shuvo, Mehedi Hasan and Jannat, Mst Raonik and Uddin, Jia and Ali, Sahara},
  journal={PLOS One},
  volume={21},
  pages={e0356383},
  year={2026},
  publisher={Public Library of Science}
}
```


The dataset itself can be cited as:

eimran. (2026). *eimran/IoT-RiceMobileNet: IoT-RiceMobileNet v1.0.0* (Version v1.0.0) [Computer software]. Zenodo. https://doi.org/10.5281/ZENODO.21140529


*This dataset was reformatted from its original format to match HuggingFace standards.*