A Microservice Architecture for Drone-Based Plant Disease Detection in Smart Agriculture

IoT-enabled technologies hold immense potential in revolutionizing smart agriculture, where they can substantially enhance crop quality and yield while reducing operational costs. Among the various applications of smart agriculture, pest control is a critical issue that demands attention. This paper presents a comprehensive fog-based smart agriculture system that integrates edge computing and UAV technology to detect a broad spectrum of diseases in tomato plants. We develop advanced deep learning models and introduce a novel algorithm for precise and efficient disease detection. Our system rapidly identifies unhealthy plants and accurately diagnoses specific diseases using a robust edge server, even in rural areas with limited or no internet connectivity. Experimental results demonstrate the system’s efficacy in detecting diseases while significantly reducing costs compared to existing state-of-theart systems.