README.md

metadata

configs:
  - config_name: raw
    default: true
    data_dir: raw
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Defect Dragon Fruit
              '1': Fresh Dragon Fruit
  - config_name: augmented
    data_dir: augmented
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Defect Dragon Fruit
              '1': Fresh Dragon Fruit
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 1K<n<10K

Dragonfruit Quality Classification

A dataset for quality classification of dragonfruit. The dataset contains raw and augmented versions.
The raw dataset contains 1,652 images.
Images per class:

• Defect Dragon Fruit: 754
• Fresh Dragon Fruit: 898

The augmented dataset contains 5,000 images.
Images per class:

• Defect Dragon Fruit: 3,000
• Fresh Dragon Fruit: 2,000

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

Citation

@article{khatun2024comprehensive,
  title={A comprehensive dragon fruit image dataset for detecting the maturity and quality grading of dragon fruit},
  author={Khatun, Tania and Nirob, Md Asraful Sharker and Bishshash, Prayma and Akter, Morium and Uddin, Mohammad Shorif},
  journal={Data in Brief},
  volume={52},
  pages={109936},
  year={2024},
  publisher={Elsevier}
}

Khatun, Tania; Nirob, Md. Asraful Sharker ; Bishshash, Prayma ; Uddin, Mohammad Shorif (2023), “Dragon Fruit Maturity Detection and Quality Grading Dataset”, Mendeley Data, V1, doi: 10.17632/2jpzbx8tm6.1