Plant Disease Detection Using Image Processing and Machine Learning

Abstract

Our Plant disease detection project presents a Convolutional Neural Network (CNN) model for the classification of plant diseases based on image data. The dataset comprises images of various plant diseases and healthy plants obtained from the "PlantVillage" database. The images are preprocessed by resizing them to a standard size and applying augmentation techniques. The CNN model is built using the Keras library and consists of multiple convolutional layers followed by pooling, batch normalization, and dropout layers. The model is trained using the Adam optimizer and evaluated on a test set. The training and validation accuracy and loss are plotted over the epochs to analyze the model's performance. The trained model achieves a certain accuracy on the test set, indicating its potential for accurately identifying plant diseases. The saved model can be utilized for real-world applications in plant disease detection and management, providing valuable assistance to farmers and researchers.