A novel deep learning and Internet of Things (IoT) enabled precision agricultural framework for crop yield production

Abstract

<p>Precision agriculture is a growing concept that frequently refers to enhancing farms via the use of up-to-date knowledge and cutting-edge technology, which in turn aids farmers by automating and improving them to increase rural profitability. This paper suggests the novel framework Deep-Plant-IoT which amalgamates the Internet of Things (IoT) and Deep learning framework for an effective prediction of crop yields which act as intelligent recommendation systems that can significantly improve the production. The framework incorporates IoT sensors and devices to collect and store the soil parameters in the cloud. Then these data are downloaded offline and the Harris Hawk Optimized Long Short Term Memory network is deployed to effectively predict crop yields that can aid in better production. Nearly 15902 data were collected for two months and Extensive testing was undertaken to employ these data to evaluate and analyze the proposed framework. Moreover, the prediction algorithm proposed in the framework is evaluated in comparison to other cutting-edge learning models. The suggested algorithm has demonstrated greater performance such that 98% accuracy, 97.23% precision, 97.0% recall, and 97.2% F1-score respectively.</p>