Irish Potato Imagery Dataset for Detection of Early and Late Blight Diseases

Abstract

This dataset comprises of 58,709 annotated images of irish potato leaves, categorized into three classes (healthy, early blight and late blight). The data was collected over six months from smallholder farms in Southern Highlands Tan- zania, using Samsung Galaxy A03 smartphones with 8- megapixel camera. Researchers, farmers and agricultural ex- tension officers were trained to capture images under di- verse conditions, including varying lighting, angles and back- grounds to ensure the dataset is diverse and representative. Plant pathologists were used to validate the images to en- sure and enhance the reliability of the labels. Pre-processing steps such as duplicate removal, filtering of irrelevant im- ages, annotation and metadata integration were applied re- sulting in a high-quality dataset. The dataset is organized into three folders (healthy, early blight and late blight) and is freely available on the Zenodo repository to promote ac- cessibility for researchers working in the field of plant dis- eases. This dataset holds significant potential for reuse in training machine learning models for crop disease detec- tion, transfer learning and data augmentation studies. By en- abling early detection and classification of potato diseases, the dataset supports the development of innovative agricul- tural tools aimed at reducing crop losses and enhancing food security in Sub-Saharan Africa. Its robust design and regional specificity make it a valuable resource for advancing research and innovation in sustainable farming practices.