---
dataset_info:
  features:
  - name: image
    dtype: image
  - name: label
    dtype:
      class_label:
        names:
          '0': anomalous
          '1': occluded
          '2': ripe
          '3': unripe
  splits:
  - name: train
    num_bytes: 28873401
    num_examples: 3520
  download_size: 27091521
  dataset_size: 28873401
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
task_categories:
- image-classification
size_categories:
- 1K<n<10K
---

# Riseholme Strawberry Classification 2021

A dataset for image classification of Riseholme Strawberry Classification 2021. The dataset contains 3,520 images across 4 classes: anomalous, occluded, ripe, unripe.  
Images per class:
- anomalous: 153
- occluded: 499
- ripe: 462
- unripe: 2,406

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

## Citation

```bibtex
@inproceedings{CWSC21,
title={Self-supervised Representation Learning for Reliable Robotic Monitoring of Fruit Anomalies},
author={Choi, Taeyeong and Would, Owen and Salazar-Gomez, Adrian and Cielniak, Grzegorz},
booktitle={2022 International Conference on Robotics and Automation (ICRA)},
pages={2266--2272},
year={2022},
organization={IEEE}
}
```

https://github.com/ctyeong/Riseholme-2021