Plant Disease Detection and Recognition Using Machine Learning

Plant Disease Detection and Recognition Using Machine Learning

Karuna Pawar, Apeksha Gangurde, Sakshi Rahane, Dr.Ranjana Dahake

Final Year Project, MET's Institute of Engineering, Department of Computer Engineering, Nashik.

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

Abstract - The agriculture industry plays a pivotal role in ensuring global food security. However, plant diseases pose a significant threat to crop yields, making early detection and diagnosis crucial. This project explores the application of machine learning, specifically the VGG-16 convolutional neural network (CNN) algorithm, for the automatic detection and recognition of plant diseases. Our study involves a comprehensive dataset of diseased and healthy plant images, representing various plant species and disease types. The proposed methodology consists of several key steps. First, a dataset of labeled plant images is curated and prepossessed. Data augmentation techniques are applied to enhance the model's generalization capabilities. The VGG-16 model is utilized for feature extraction, leveraging its deep learning architecture, which has demonstrated remarkable performance in image classification tasks. Through extensive training and fine-tuning, the VGG-16-based model can train to accurately classify plant images into healthy or diseased categories. Evaluation metrics, such as accuracy, precision, recall, and F1-score, are used to assess the model's performance. The results will showcase the effectiveness of the VGG-16 algorithm in differentiating between healthy and diseased plants, contributing to early disease detection and reducing yield losses.

Key Words:  Machine Learning, VGG-16, Disease Detection ,Image Processing