Enhancing Agricultural Diagnostics

Neural Network-Based Plant Disease Identification

Abstract

This study aims to develop a machine-learning model for the automated identification of plant diseases using leaf images. The analysis uses the PlantVillage dataset and utilizes a multi-layer perceptron (MLP) neural network to classify images from the PlantVillage dataset containing various plant species and diseases. Performance was evaluated using accuracy, F1 score, and confusion matrices. The model achieved an accuracy of 88.06% in distinguishing between healthy and diseased plants when optimized with the appropriate hyperparameters. These results highlight the potential of neural networks as tools for disease detection in agriculture, offering a scalable and reliable solution that can significantly enhance crop management practices. Future work could explore incorporating additional diverse datasets and advanced neural network models to further improve model functionality and generalization.