---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': Broken
          '1': Immature
          '2': Intact
          '3': Skin-damaged
          '4': Spotted
  splits:
  - name: train
    num_bytes: 89884943
    num_examples: 5513
  download_size: 90329805
  dataset_size: 89884943
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: cc-by-4.0
task_categories:
- image-classification
size_categories:
- 1K<n<10K
---
# Soybean Damage Classification

A dataset for damage classification of soybean seeds. The dataset contains 5,513 images across 5 classes: Broken, Immature, Intact, Skin-damaged, Spotted.  
Images per class:
- Broken: 1,002
- Immature: 1,125
- Intact: 1,201
- Skin-damaged: 1,127
- Spotted: 1,058

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

## Citation

```bibtex
@article{lin2023soybean,
  title={Soybean image dataset for classification},
  author={Lin, Wei and Fu, Youhao and Xu, Peiquan and Liu, Shuo and Ma, Daoyi and Jiang, Zitian and Zang, Siyang and Yao, Heyang and Su, Qin},
  journal={Data in Brief},
  volume={48},
  pages={109300},
  year={2023},
  publisher={Elsevier}
}
```

Lin, Wei; Fu, Youhao; Xu, Peiquan; Liu, Shuo; Ma, Daoyi; Jiang, Zitian; zang, siyang; Yao, Heyang; Su, Qin (2023), “Soybean Seeds”, Mendeley Data, V6, doi: 10.17632/v6vzvfszj6.6

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