Dataset Viewer
The dataset viewer is not available for this subset.
Cannot get the split names for the config 'default' of the dataset.
Exception:    SplitsNotFoundError
Message:      The split names could not be parsed from the dataset config.
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 286, in get_dataset_config_info
                  for split_generator in builder._split_generators(
                                         ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      StreamingDownloadManager(base_path=builder.base_path, download_config=download_config)
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/webdataset/webdataset.py", line 84, in _split_generators
                  raise ValueError(
                  ...<2 lines>...
                  )
              ValueError: The TAR archives of the dataset should be in WebDataset format, but the files in the archive don't share the same prefix or the same types.
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/split_names.py", line 68, in compute_split_names_from_streaming_response
                  for split in get_dataset_split_names(
                               ~~~~~~~~~~~~~~~~~~~~~~~^
                      path=dataset,
                      ^^^^^^^^^^^^^
                      config_name=config,
                      ^^^^^^^^^^^^^^^^^^^
                      token=hf_token,
                      ^^^^^^^^^^^^^^^
                  )
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 340, in get_dataset_split_names
                  info = get_dataset_config_info(
                      path,
                  ...<6 lines>...
                      **config_kwargs,
                  )
                File "/usr/local/lib/python3.14/site-packages/datasets/inspect.py", line 291, in get_dataset_config_info
                  raise SplitsNotFoundError("The split names could not be parsed from the dataset config.") from err
              datasets.inspect.SplitsNotFoundError: The split names could not be parsed from the dataset config.

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

Pepper Peduncle Detection Dataset

This is the cropped pepper and peduncle instance-segmentation dataset used in HortiBot: An Adaptive Multi-Arm System for Robotic Horticulture of Sweet Peppers (IROS 2024).

Dataset Description

The dataset contains pepper-centered image crops and YOLO polygon segmentation labels for two classes:

ID Class
0 peduncle
1 pepper
Split Image/label pairs Peduncle instances Pepper instances
Train 94,077 49,615 94,195
Validation 10,805 4,118 10,806
Total 104,882 53,733 105,001

The split archives expand into the Ultralytics YOLO segmentation layout:

archives/
  train.tar.gz
  val.tar.gz
data.yaml
models/
  best.pt

Each non-empty label line contains a class identifier followed by normalized polygon coordinates. The checkpoint is the matching YOLOv8L segmentation model trained with 96 px inputs.

Extract both archives into one directory, then use data.yaml from this repository:

tar -xzf train.tar.gz
tar -xzf val.tar.gz

Archive SHA-256 checksums:

0881eafba72652190b2fb6470d0b10f5202fec29de09634939f999faeb8a06cd  train.tar.gz
a79ebe3967eefe6e30639410cda9c6bd02a8ae62d68f2cb19d33aa00e015f834  val.tar.gz

Intended Use

The data supports research on instance segmentation of sweet peppers and their peduncles in pepper-centered crops. In HortiBot, a full-image pepper detector first identifies each fruit, then the expanded fruit crop is passed to this model. The dataset is not a benchmark for unrestricted full-image plant-organ detection.

Source Data and Licensing

This release is a derived dataset combining crops from:

  1. Synthetic and Empirical Capsicum Annuum Image Dataset, R. Barth, DOI: 10.4121/uuid:884958f5-b868-46e1-b3d8-a0b5d91b02c0.
  2. Sweet Pepper and Peduncle Segmentation, Luis Enrique Montoya Cavero and Jesús Escobedo, distributed under CC BY-SA 4.0 at Kaggle.
  3. BUP20 Sweet Pepper Dataset, described on the University of Bonn AgRobotics dataset page.

The files retain the terms of their respective source datasets. This combined release is therefore marked other rather than applying a new blanket license. Users are responsible for reviewing and complying with each source dataset's terms. In particular, the BUP20 source page specifies citation requirements but does not state a standard license.

Limitations

  • The dataset includes augmented variants derived from the same source images; image counts must not be interpreted as independent acquisition events.
  • The train/validation split is supplied as used by the project and is not guaranteed to separate all augmented variants by original source image.
  • The imagery is specific to sweet-pepper plants and the source acquisition or rendering conditions. Performance outside those conditions was not evaluated by this dataset release.
  • Three training labels are empty background examples.
  • images/train/synthetic_image_color_84110.jpg is only 96 x 8 pixels and is reported as corrupt by Ultralytics, which skips it during training.

Paper Results

The HortiBot paper reports a peduncle mask mAP@50 of 0.741 and an F1 score of 0.781. Pepper and peduncle detection together required 100 ms on the HortiBot system. These are reported paper results, not metrics recomputed during this release.

Citation

@inproceedings{lenz2024hortibot,
  author={Lenz, Christian and Menon, Rohit and Schreiber, Michael and Jacob, Melvin Paul and Behnke, Sven and Bennewitz, Maren},
  title={{HortiBot}: An Adaptive Multi-Arm System for Robotic Horticulture of Sweet Peppers},
  booktitle={2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2024},
  pages={31--38},
  doi={10.1109/IROS58592.2024.10802082}
}

Please also cite the source datasets relevant to your use.

Downloads last month
41