README.md

metadata

configs:
  - config_name: default
    default: true
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': 1st-grade
              '1': 2nd-grade
              '2': 3rd-grade
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 1K<n<10K

AFruitDB Fruit Grade Classification

A dataset for grade classification of 6 fruits: tomato, papaya, mango, Burmese grape, apple, and banana. The dataset contains 3,167 images across 3 classes: 1st-grade, 2nd-grade, 3rd-grade. Images per class:

• 1st-grade: 1,297
• 2nd-grade: 1,196
• 3rd-grade: 674

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

Citation

@article{mojumdar2025afruitdb,
  title={AFruitDB: A comprehensive dataset of six commonly used Asian fruits for advanced grading and biodiversity insights},
  author={Mojumdar, Mayen Uddin and Islam, Shahrin and Al Mamun, Md and Hasan, Rifat and Siddiquee, Shah Md Tanvir and Chakraborty, Narayan Ranjan},
  journal={Data in Brief},
  volume={59},
  pages={111380},
  year={2025},
  publisher={Elsevier}
}

Mojumdar, Mayen Uddin ; Mamun, Md Al ; Islam, Shahrin; Hasan, Rifat (2024), “A Dataset of Common Asian Fruits for Quality Grading and Biodiversity Research”, Mendeley Data, V1, doi: 10.17632/bz65dz2pbj.1