An IoT-based cognitive monitoring system for early plant disease forecast

In this paper, we develop an IoT-based monitoring system for precision agriculture applications such as epidemic disease control. Such an agricultural monitoring system provides environmental monitoring services that maintain the crop growing environment in an optimal status and early predicts the conditions that lead to epidemic disease outbreak. The agricultural monitoring system provides a service to store the environmental and soil information collected from a wireless sensor network installed in the planted area in a database. Furthermore, it allows users to monitor the environmental information about the planted crops in real-time through any Internet-enabled devices. We develop artificial intelligence and prediction algorithms to realize an expert system that allows the system to emulate the decision-making ability of a human expert regarding the diseases and issue warning messages to the users before the outbreak of the disease. Field experiments showed that the proposed system reduces the number of chemical applications, and hence, promotes agriculture products with no (or minimal) chemicals residues and high-quality crops. This platform is designed to be generic enough to be used with multiple plant diseases where the software architecture can handle different plant disease models or other precision agriculture applications.