Design of BQN-based decision support system and HSCNN-GPOR futuristic prediction for precision agriculture using IoT

Abstract Precision agriculture (PA) makes use of technology to increase sustainability by making better use of resources like land, water, fuel, fertilizer, and pesticides. In PA, crop productivity and profitability are impacted by weather, fertilizer usage, pests, crop irrigation, and crop diseases. In many existing techniques, there is no effective irrigation control and futuristic prediction of diseases in crops and soil fertility. To overcome this issue, a novel Bayesian Q-Network (BQN)-based decision support system and Hopfield symmetric convolutional neural network-Gaussian probabilistic ordinal regression (HSCNN-GPOR) have been proposed, for effective decision-making and futuristic prediction and classification of diseases in crops. In this technique, a decisive logistic associative rule-based BQN model has been introduced to overcome the weak integrated decisions and an inter-temporal comparison of soil fertility and environmental information. This technique is used to obtain the soil fertility and to improve the intertemporal characteristics by Fuzzy logistic associative rule to obtain environmental information. In this existing model, to make a decision-making process, BQN has been introduced, to improve the intertemporal characteristics to make an effective decision for irrigation control and fertilizer usage. Moreover, the existing techniques predict crop diseases and pests from collected crop images, but they do not predict the future condition of crops from these crop images. Hence, a model Hopfield Symmetric CNN-GPOR has been introduced to classify pests and predict future diseases in crops. In this model, a symmetric Hopfield CNN extracts features from crop images, which are then utilized by Gaussian Probabilistic Ordinal Regression to classify crop diseases and pests, along with future predictions. Therefore, the proposed model provides the ability to make wise decisions, futuristic predictions of diseases in crops, and classify the pests in crops with high recognition, accuracy, and classification rates.