Early Detection and Classification of Potato Leaf Disease Using an Efficient Deep Learning Model

Early Detection and Classification of Potato Leaf Disease Using an Efficient Deep Learning Model

Author(s): Samiksha Madan More, R Dileep Kumar, Hari Mohan Singh

Affiliation(s): Department of Computer Science and Information Technology,

Vaugh Institute of Agricultural Engineering And Technology,

Sam Higginbottom University of Agriculture, Technology And Sciences, Prayagraj,

U.P, INDIA 211007

Corresponding Author: sameekshamorey@gmail.com,

Abstract:

Plant diseases impact the availability and safety of plants for human and animal consumption, as well as the safety of food, limiting food availability and access and crop output and quality. Especially plants that contribute to the national economy as well as the food needs of its own people. Manual diagnosis of disease in early stages requires so much of expertise, technology assisted automatic disease classification approaches are becoming new norm now a days. In this study, an effort is made to detect potato leaf disease as India is second largest country producing potato. In this research a novel deep learning model for potato leaf disease classification has developed using CNN models and tested on PlantVillage dataset. The proposed model is trained to accurately classify potato leaf disease in 3 classes & the method achieved 99.61% accuracy. The outcome of the experiment demonstrates that, the proposed model outperforms pre-trained model i.e. VGG16 and InceptionV3. The proposed method also tested with respect to its consistency and reliability.

Keywords: potato leaf disease classification, deep learning, crop health, convolutional neural network, supervised learning