Project-AgML/taro_blight_stage_classification

Taro Blight Stage Disease Classification

A dataset for image classification of Taro blight stage. The dataset contains 18,248 images across 4 classes: Early_Blight, Healthy, Late_Blight, Mid_Blight. Images per class:

• Early_Blight: 4,864
• Healthy: 8,744
• Late_Blight: 1,270
• Mid_Blight: 3,370

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

Citation

@article{nwaneto2025image,
  title={Image dataset of Taro Leaf Blight disease collected from the West African Sub-Region},
  author={Nwaneto, Chidiebere and Yinka-Banjo, Chika and Ugot, Ogban-Asuquo and Umeugochukwu, Obiageli and Annor, Thompson},
  journal={Data in Brief},
  pages={111869},
  year={2025},
  publisher={Elsevier}
}```

Yinka-Banjo, Chika; Nwaneto, Chidiebere; Ugot, Ogban-Asuquo; Umeugochukwu, Obiageli; Annor, Thompson (2024), “An Image Dataset of Taro Leaf Blight Disease Collected from the West African Sub-Region”, Mendeley Data, V2, doi: 10.17632/3knm93dkc5.2