Mulberry Leaf Disease Detection System

The early detection of plant diseases is critical for minimizing crop damage and ensuring sustainable agricultural practices. In the case of mulberry plants, which are essential for the sericulture industry, diseases such as leaf spot, leaf rust, and powdery mildew pose significant threats to crop health and productivity. Traditional disease detection methods are often inefficient, costly, and laborintensive, making it difficult for farmers to identify and manage infections in a timely manner. This project presents a novel approach to automating mulberry leaf disease detection using deep learning techniques. The system utilizes a combination of two widely recognized convolutional neural networks (CNNs): ResNet50 and VGG19. These models are employed to extract rich feature representations from leaf images and are then fused to improve classification accuracy. The integration of these models allows for the detection of multiple disease types with greater robustness and precision compared to using a single model alone. The proposed solution is designed to be part of a mobile system capable of navigating through mulberry fields, capturing leaf images, and performing real-time disease analysis. The system is integrated with the human interface software with the disease identification system programmed to identify the disease. This system's performance is evaluated based on metrics such as accuracy, precision, and recall, demonstrating its effectiveness in detecting the targeted leaf diseases. By leveraging the power of deep learning, this project aims to offer a cost-effective, scalable tool for farmers to protect their crops and increase yields. Ultimately, the implementation of this system could lead to healthier crops, reduced pesticide usage, and a more sustainable approach to agriculture.