Datasets:

Project-AgML
/

jute_disease_classification

configs:
  - config_name: default
    default: true
    features:
      - name: image
        dtype: image
      - name: label
        dtype:
          class_label:
            names:
              '0': Dieback
              '1': Fresh
              '2': Holed
              '3': Mosaic
              '4': Stem Soft Rot
license: cc-by-4.0
task_categories:
  - image-classification
size_categories:
  - 1K<n<10K

Jute Disease Classification

A dataset for image classification of Jute Disease. The dataset contains 1,390 images across 5 classes: Dieback, Fresh, Holed, Mosaic, Stem Soft Rot. Images per class:

Dieback: 300
Fresh: 280
Holed: 300
Mosaic: 240
Stem Soft Rot: 270

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

Citation

@article{islam2025comprehensive,
  title={A comprehensive image dataset of jute diseases},
  author={Islam, Md Masudul and Sheikh, Md Ripon},
  journal={Data in Brief},
  pages={112334},
  year={2025},
  publisher={Elsevier}
}

Islam, Md. Masudul; Sheikh, Md Ripon, 2025, "Jute Disease Image Dataset", https://doi.org/10.7910/DVN/FJ1DM1, Harvard Dataverse, V1