Evaluation of Current IoT Approaches in Cereal Crop Pests and Diseases Management

Abstract

The inevitable increase in the world population proliferates the imposition of more food production. Moreover, the series of insurgency activities in rural areas causes a frequent reduction in the workforce causing Food manufacturing to face difficulties due to a rise in production costs in this part of the world. Most farms lack management due to abandonment by the owner for fear of being attacked, making the crops to be prone to different management risks including the invasion of pests and diseases. The current issues of food production can be overcome with smart farming, which uses the Internet of Things (IoT) concept for farm management. This research effort evaluates the current research work in which IoT approaches were used in crops pests and diseases management, using the Preferred Reporting Items for Systematic Reviews (PRISMA) methodology. The research work aims at improving on the setbacks encountered with the previous approaches engaging the combination of IoT and artificial intelligence; using the technology of deep learning which is more stable, more secure, and with higher recognition accuracy. This proposed research work makes use of the Long Short Term Memory (LSTM) architecture with Association rule analysis using the Apriori algorithm to mine the association rule to enhance pests and diseases segmentation and recognition systems.
The result shows how data processing practices have changed recently. The proposed system displayed an impressing accuracy rate and a negligible error rate.