Intelligent IoT Environment for Early Crop Disease Prediction and Management

Early crop disease prediction in agriculture is crucial to increasing profitability through efficient production, post-harvest, and marketing management. Early crop disease prediction is necessary in agriculture for efficient productivity and sustainability. Manual crop disease prediction is prone to errors and time-consuming because it mainly depends on manual examination and laboratory analysis. We use Recurrent Neural Networks (RNNs) to interpret the data received from sensors, and image analysis is done using Convolutional Neural Networks (CNNs). The system developed by integrating the Internet of Things (IoT) and Deep Learning (DL) is efficient in detecting crop disease. The proposed crop disease detection system provides $\mathbf{9 9 \%}$ accuracy, that is superior to other existing methods. Additionally, the proposed system provides useful insights to farmers, early intervention in crop disease prediction, and minimizes the transmission of disease among the crops. Thus, the proposed system utilizes the resources efficiently and reduces crop losses.