Leaf Disease Detection and Prevention using Deep Learning

Read full text Summarise with VIRAT
Authors: G. Sirisha, D Shishira, B. Surya Shekar, C. Naveen Kumar, B Charan
Cited by: 0
References: 8

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

Global agricultural productivity is increasingly threatened by climate variability, plant diseases, and the need for sustainable crop management. Among major crops, tomatoes are widely cultivated and are highly susceptible to a variety of diseases caused by bacteria, fungi, viruses, and pests. Early and accurate detection of these diseases is essential to minimize crop loss and improve yield. This paper presents an AI-based Plant Leaf Disease Detection and Prevention System that leverages deep learning and image classification techniques to identify tomato leaf diseases and recommend appropriate preventive measures. The proposed system utilizes transfer learning with convolutional neural networks (CNNs) implemented using TensorFlow. Leaf images are directly used as input without requiring complex preprocessing or manual feature extraction. The trained model is capable of classifying ten categories, including nine common tomato diseases—bacterial spot, early blight, late blight, leaf mold, septoria leaf spot, spider mites, target spot, yellow leaf curl virus, and tomato mosaic virus—along with the healthy condition. A user-friendly interface is developed using Gradio, allowing users to upload or capture leaf images for real-time prediction. The system outputs the top three predicted disease classes along with confidence scores, identifies the most probable disease, and provides corresponding prevention strategies and remedies. This integrated approach not only detects diseases but also supports decision-making for effective crop management. Experimental evaluation demonstrates high classification performance, achieving over 99% accuracy, precision, recall, and F1-score. The system is designed to be deployable on standard computing devices such as laptops, making it accessible to farmers and agricultural experts. Overall, the proposed solution offers a cost-effective, accurate, and scalable approach for smart agriculture, enabling early disease detection and timely intervention to enhance crop productivity and sustainability.

Fields of Study & Subjects

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

Publication details

Journal / source
International Journal of All Research Education & Scientific Methods
Volume
14
Issue
03
Pages
1175–1182
Publication date
1 January 2026
Language
EN