Datasets:

Project-AgML
/

pepper_disease_pest_classification

dataset_info:
  features:
    - name: image
      dtype: image
    - name: label
      dtype:
        class_label:
          names:
            '0': Alternaria_Boltch
            '1': Brown_Spot
            '2': Grey_spot
            '3': Mosaic
            '4': Rust
  splits:
    - name: train
      num_bytes: 574217137
      num_examples: 26377
  download_size: 886186963
  dataset_size: 574217137
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 10K<n<100K

Pepper Disease Pest Classification

The dataset comprises real RGB images of pepper plants in field environments, collected over two years (2022-2023) in Shouguang City, China. Captured using a fixed platform setup, the images provide consistent visual data for computer vision research focused on agricultural disease detection in pepper crops. The dataset contains 26,377 images across 5 classes: Alternaria_Boltch, Brown_Spot, Grey_spot, Mosaic, Rust.
Images per class:

Alternaria_Boltch: 5,343
Brown_Spot: 5,655
Grey_spot: 4,810
Mosaic: 4,875
Rust: 5,694

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

Citation

@article{wang2025advanced,
  title={An advanced deep learning method for pepper diseases and pests detection},
  author={Wang, Xuewei and Liu, Jun and Chen, Qian},
  journal={Plant Methods},
  volume={21},
  pages={70},
  year={2025},
  publisher={BioMed Central}
}

liu, jun (2024), “vegetable disease”, Mendeley Data, V1, doi: 10.17632/tg3z7xxkdb.1

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