---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': cracked
          '1': crushed
          '2': no_buds
          '3': no_damage
          '4': single_damaged_buds
          '5': two_buds
  splits:
  - name: train
    num_bytes: 2301208454
    num_examples: 153
  download_size: 2301235850
  dataset_size: 2301208454
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- image-classification
size_categories:
- n<1K
---

# Sugarcane Damage Usa

A dataset for classification of sugarcane damage. The dataset contains 153 images across 6 classes: cracked, crushed, no_buds, no_damage, single_damaged_buds, two_buds.  
Images per class:
- cracked: 24
- crushed: 24
- no_buds: 21
- no_damage: 36
- single_damaged_buds: 24
- two_buds: 24

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

## Citation

```bibtex
@ARTICLE{8412587,
author={Alencastre-Miranda, Moises and Davidson, Joseph R. and Johnson, Richard M. and Waguespack, Herman and Krebs, Hermano Igo},
journal={IEEE Robotics and Automation Letters},
title={Robotics for Sugarcane Cultivation: Analysis of Billet Quality using Computer Vision},
year={2018},
volume={3},
number={4},
pages={3828-3835},
doi={10.1109/LRA.2018.2856999}
}
```