CNN-based Plant Disease Classification and Severity Analysis using Image Processing

Abstract

This paper introduces a Convolutional Neural Network (CNN)-based methodology in order to categorize plants and the analysis of their severity. The suggested model is trained using images of plant leaves so that it can accurately sort them into different disease groups. To enhance the quality of the input, image preprocessing techniques are employed and image processing techniques are used to figure out how bad the infection is by looking at tthe proportion of diseased leaf area to the entire leaf area. There are several convolutional and pooling layers in the CNN architecture. After that, there is a categorization layer that is fully connected. The model is trained with an 80:20 split train-test and data enhancement methods to make it better at generalizing and dealing with class imbalance. Experimental results show that the proposed method is more accurate than traditional machine learning models, with an accuracy of 90–95%, which is better than traditional machine learning models like Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Decision Tree (DT). The system works well and can be expanded for use in real-time farming.