Modeling of Sensor Enabled Irrigation Management for Intelligent Agriculture Using Hybrid Deep Belief Network

Abstract

Artificial intelligence (AI) technologies and sensors have recently received significant interest in intellectual agriculture. Accelerating the application of AI technologies and agriculture sensors in intellectual agriculture is urgently required for the growth of modern agriculture and will help promote smart agriculture. Automatic irrigation scheduling systems were highly required in the agricultural field due to their capability to manage and save water deficit irrigation techniques. Automatic learning systems devise an alternative to conventional irrigation management through the automatic elaboration of predictions related to the learning of an agronomist. With this motivation, this study develops a modified black widow optimization with a deep belief network-based smart irrigation system (MBWODBN-SIS) for intelligent agriculture. The MBWODBN-SIS algorithm primarily enables the Internet of Things (IoT) based sensors to collect data forwarded to the cloud server for examination purposes. Besides, the MBWODBN-SIS technique applies the deep belief network (DBN) model for different types of irrigation classification: average, high needed, highly not needed, and not needed. The MBWO algorithm is used for the hyperparameter tuning process. A wideranging experiment was conducted, and the comparison study stated the enhanced outcomes of the MBWODBN-SIS approach to other DL models with maximum accuracy of 95.73%.