Intelligent water management via deep-IoT enabled precision agriculture

Smart agriculture utilises IoT platform to enrich the quality and quantity of cultivation crops. Due to the less availability of water content precision irrigation has been introduced in agriculture. Precision irrigation needs full monitoring of agricultural regions and crops to increase the productivity of the farmland with efficient water usage. In nature due to climate change, the prediction of agriculture land characteristics is not always stable and it causes unpredictable losses. To withstand against these unstable conditions this research work introduces a novel smart deep learning (DL)-based Balancer (SDB) with remote irrigation control system. The proposed SDB framework uses a convolutional neural network with bidirectional long short-term memory network (CNN-BiLSTM) to perform categorisation and prediction based on the extracted parameters that have been received from the sensors. The proposed SDB framework has been evaluated with four existing systems such as fog-based IoT, SWS, DMVO and AgriSegNet in terms of accuracy, specificity and F1-score. The proposed SDB framework achieves higher accuracy of 26.66%, 10.76%, 17.43% and 10.76% than Fog-based IoT, SWS, DMVO and AgriSegNet.