Leaf disease detection using Matlab and deep learning

This project presents a comprehensive approach to leaf disease detection and classification using advanced image processing and machine learning techniques. The system begins by preprocessing an input leaf image through resizing and noise reduction to enhance clarity. It then applies Convolutional Neural Networks (CNN) with a pre-trained VGG16 model for leaf classification, distinguishing between normal and abnormal leaves. Abnormal leaves undergo further analysis, where segmentation techniques are utilized to identify and classify the disease. The system trains a CNN on a dataset of leaf images, refining the VGG16 model to classify specific diseases and recommend appropriate treatments based on the detected condition. Upon classification, the system automatically sends an email with the disease type, its stage, and recommended pesticides to a specified address, ensuring timely and accurate information for disease management. This integrated approach aims to improve agricultural practices by leveraging image processing and machine learning to provide real-time diagnostics and actionable recommendations.

Keywords: Leaf Disease Related Dataset, Deep Learning, Convolution Neural Network, Image Processing Techniques, segmentation and accuracy.