Datasets:

Project-AgML
/

raspberry_maturity_detection

dataset_info:
  features:
    - name: image
      dtype: image
    - name: objects
      struct:
        - name: bbox
          list:
            list: float64
        - name: categories
          list:
            class_label:
              names:
                '0': Buds
                '1': Flowers
                '2': Unripe Berries
                '3': Ripe Berries
                '4': Damaged Buds
  splits:
    - name: train
      num_bytes: 1537146621
      num_examples: 2039
  download_size: 1545022288
  dataset_size: 1537146621
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - object-detection
size_categories:
  - 1K<n<10K

Raspberry Maturity Detection

This dataset comprises real-world RGB images capturing raspberry plants in field conditions across multiple maturity stages. Collected during June-August 2021 using handheld Apple iPhone XS devices in Latvia, it provides diverse environmental lighting and plant growth scenarios for object detection research. The dataset contains 2,039 images with 46,656 bounding box annotations across 5 categories.

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

Citation

<!-- TODO: add BibTeX citation -->

The dataset itself can be cited as:

Strautiņa, S., Kalniņa, I., Kaufmane, E., Sudars, K., Namatēvs, I., Ņikuļins, A., Judvaitis, J., & Balašs, R. (2022). RaspberrySet: Dataset of Annotated Raspberry Images for Object Detection. Zenodo. https://doi.org/10.5281/ZENODO.7014728

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