DeepQ Based Automated Irrigation Systems Using Deep Belief WSN

Abstract

Deep learning is the subset of artificial intelligence and it is used for effective decision making. Wireless Sensor based automated irrigation system is proposed to monitor and cultivate crop. Our system consists of Distributed wireless sensor environment to handle the moisture of the soil and temperature levels. It is automated process and useful for minimizing the usage of resources such as water level, quality of the soil, fertilizer values and controlling the whole system. The mobile app based smart control system is designed using deep belief network. This system has multiple sensors placed in agricultural field and collect the data. The collected transmitted to cloud server and deep learning process is applied for making decisions. DeepQ residue analysis method is proposed for analyzing automated and sensor captured data. Here, we used 512 × 512 × 3 layers deep belief network and 10000 trained data and 2500 test data are taken for evaluations. It is automated process once data is collected deep belief network is generated. The performance is compared with existing results and our process method has 94% of accuracy factor. Also, our system has low cost and energy consumption also suitable for all kind of agricultural fields.