DESIGN AND DEVELOPMENT OF AGRI BOT FOR SEEDING, PLOUGHING AND WATERING PURPOSE

This study focuses on leaf disease classification utilizing Convolutional Neural Network (CNN) and proposes an approach for suggesting appropriate fertilizers based on the identified diseases. The process involves taking input images of affected leaves and subjecting them to pre-processing steps, including image resizing, restoration, and noise removal. Subsequently, a CNN, a deep learning algorithm, is employed for disease classification. The identified leaf diseases encompass Cherry Powdery Mildew, Corn Cercospora Leaf Spot, Gray Leaf Spot, Peach Bacterial Spot, Potato Early Blight, Strawberry Leaf Scorch, and Tomato Mosaic Virus. Furthermore, the study extends its scope by recommending specific fertilizers tailored to each identified disease, aiming to enhance plant health and mitigate the impact of these diseases. The model's accuracy in disease classification is assessed, contributing to the evaluation of its practical utility in agricultural settings. This holistic approach intertwining disease identification, classification, and fertilizer recommendation showcases the potential of advanced technologies in precision agriculture, offering targeted solutions for crop management and fostering improved yield outcomes.

Keywords: Leaf Disease, pre-processing, convolutional neural networks, Deep learning technique, Fertilizers and Accuracy.