Development of AI-Based Maize Storage Monitoring System Using IoT

The agriculture sector in Rwanda put much consideration into maize farming on which the acceleration in sustainable agriculture and rural development is being based, as maize has a high yield compared to other historically produced grains. It is relatively easy to conserve, and serves as a significant source of income for farmers. However, there is still a poor maize storage monitoring approach, which leads to a loss of a considerable amount of maize every year. To effectively maintain and overcome grain damage, this research proposes an IoT-based maize storage monitoring system composed of sensors that detect the environmental parameters in the maize storing space and a GSM module that avails the sensed information to the server for database storage using the GPRS protocol. In addition, a mobile application built using Flutter / Dart language has been used, it comprises several interfaces that avail the status of environmental conditions in maize storing space to the user on an interval of 10 seconds matching the configuration of the GSM. A machine learning model has been developed where AI maize image classification has been made to help the identification of the maize status based on the picture of the maize taken.