Advancements in agricultural technology are increasingly focusing on the integration of automation and artificial intelligence to improve crop monitoring and disease management. Among the various innovations, Unmanned Aerial Vehicles (UAVs), commonly known as drones, are playing a vital role in modernizing field data acquisition. When coupled with Machine Learning (ML) models, UAVs become powerful tools for the real-time detection of leaf diseases, which are a major cause of yield reduction globally. This research presents an integrated approach to using drone-captured imagery and deep learning-based classification algorithms to detect and categorize plant leaf diseases across large agricultural fields. Using RGB and multispectral cameras mounted on drones, high-resolution images are captured and processed to detect anomalies in leaf structures, color, and texture. The study implements a Convolutional Neural Network (CNN) for image classification, yielding high precision and recall. Experimental results demonstrate a classification accuracy of over 95%, highlighting the model's potential for deployment in real-world farming conditions. This system drastically reduces the need for manual scouting and enables early detection and mitigation of plant diseases, leading to optimized pesticide usage and improved crop health management. The proposed solution is particularly beneficial for resource-constrained farmers as it offers a cost-effective and scalable plant health monitoring alternative. The research concludes by discussing possible enhancements using edge computing, real-time mobile integration, and expansion to pest and soil health monitoring in future work.