Web App – Based Solution to Identify Disease in Plants Using CNN

Web App - Based Solution to Identify Disease in Plants Using CNN

Gopal Singh1, Anjali Nagre2, Tanvi Sakpal3, Sonal Banpatte4, Prof. Minal Chaudhari5

1,2,3,4,5 COMPUTER ENGINEERING, INDALA COLLEGE OF ENGINEERING, KALYAN

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - This paper presents a deep learning- based    an

 approach for plant disease detection using the VGG19 Convolutional Neural Network (CNN) model. The study focuses on identifying diseases in potato, Tomato, Corn, pepper and banana crops, leveraging a dataset comprising 12 distinct cases, including 10 diseased and 2 healthy conditions. The VGG19 model, fine-tuned for this task, achieved a test accuracy of 93.50%, demonstrating its effectiveness in accurately classifying plant diseases. Additionally, image-processing techniques were integrated to estimate disease severity by quantifying the percentage of infected areas on plant leaves. The proposed system offers a scalable and efficient solution for early disease detection, enabling farmers to take timely corrective actions. By combining deep learning with advanced image analysis, this research contributes to sustainable agricultural practices, improved crop management, and enhanced food security. The findings highlight the potential of AI-driven tools in transforming plant disease management and supporting global agricultural productivity.

Key words: Agricultural productivity, convolutional neural networks (CNN), deep learning, image processing plant disease detection, VGG19.