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

Published: 25 April 2024| Version 1 | DOI: 10.17632/bd553pdtny.1


The image of the pigeonpea leaves was collected from cultivation site in the location Agriculture College, Hittinahalli Campus, Vijayapur (Lat 16.769281 and Long 75.748891), Karnataka. Visible light images of pigeonpea leaves were captured in natural conditions with various angles using Sony Cyber-Shot DSCW810 Digital Camera and Smartphone Oppo F19 pro camera. The dataset includes both healthy and infected with Cercospora Leaf Spots, Leaf Webber and Sterilic Mosaic diseases. The dataset contains 1,000 .jpg format images in 256X256 size. Data were uploaded to the repository in 4 distinct folders: 1 folder for healthy data contains 196 images, 1 folder for Cercospora Leaf spot consist of 336 images , 1 folder for Leaf webber contains 146, 1 folder for Sterilic mosaic diseases consist of 292 images so there are total 1000 images in dataset. In addition, each folder's name reflected the associated image class. The Healthy folder comprises a collection of images depicting pigeon pea leaves in a healthy condition. The Cercospora Leaf Spot folder contains images of pigeonpea leaves affected by Cercospora Leaf spot. The folder titled Leaf webber contains images depicting pigeonpea leaves that have been infected by Leaf webber. The Sterilic mosaic folder comprises images of pigeonpea leaves infected by Sterilic mosaic disease. The aim of collecting the dataset is to design an automated Pigeonpea plant leaf disease identification and classification systems based on Computer Vision algorithms.



Akkamahadevi Women's University


Agricultural Science, Computer Vision, Image Processing, Plant Pathology