A novel potato leaf dataset for detection and classification of potato leaf diseases

Read full text Summarise with VIRAT
Authors: Buddhadev Sasmal, Suparna Biswas, Ramesh Saha, Krishna Gopal Dhal, Ruba Abu Khurma, Akash Das, Anuvab Sahu, Raja Das, Bubun Pal, Kaushik Mazumdar
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The image of the potato leaves was collected from the agricultural land of Paikarapur village in the Paschim Medinipur district of West Bengal, India, Pin-721305, which is both healthy and infected with early leaf spot and late leaf spot. The dataset contains 12,000 jpg images in 2500 × 2500 pixel format, each labeled with the image number and the name of the image. Data were uploaded to the repository in 3 distinct folders: 1 folder for healthy leaves, 1 folder for early leaf spot, and 1 folder for late leaf spot diseases. In addition, each folder's name reflected the associated image class. The HEALTHY folder comprises a collection of images depicting potato leaves in a healthy condition. The EARLY LEAF SPOT folder contains images of potato leaves affected by early leaf spot. The LATE LEAF SPOT folder contains images of potato leaves infected by late leaf spot. The images were divided into distinct directories to facilitate the process of sharing and downloading data.

Fields of Study & Subjects

Smart Agriculture and AI Advanced Neural Network Applications Industrial Vision Systems and Defect Detection

Publication details

Journal / source
Mendeley Data
Publication date
30 March 2026
Language
EN
Document type
Dataset