IoT Enabled Soil Moisture and Heat Level Prediction Using Chimp Shuffled Shepherd Optimization-Based Deep LSTM for Plant Health Monitoring

Abstract

Plant health monitoring is a very significant task in any agriculture-based environment. The Internet of Things (IoT) plays an important role in the monitoring of plant diseases. IoT is required to obtain data through sensor nodes for finding soil moisture and heat level. Even though different methods are available to monitor the health of plants, observing heat level and soil moisture still results a complex task. Thus, this paper introduces a novel chimp shuffled shepherd optimization (ChSSO) by the integration of chimp optimization algorithm (ChOA) and shuffled shepherd optimization (SSOA) to perform the selection of cluster head (CH) and routing process. The proposed ChSSO is trained using the deep LSTM which is developed for predicting soil moisture and heat level conditions in IoT network to monitor the health of plants. The proposed method obtained higher performance by the metrics, like testing accuracy and precision of 0.937, and 0.926 for 100 nodes and the values of 0.940, and 0.940 for 150 nodes using the LDAS dataset.