Intelligent Crop Disease Prediction Using Iot Sensors And Deep Learning

Abstract

Intelligent crop disease prediction has become an essential component of precision agriculture, enabling early disease detection and minimizing crop losses. However, accurately identifying crop diseases remains challenging due to varying environmental conditions, disease symptoms, and the complexity of agricultural ecosystems. This research proposes an intelligent crop disease prediction framework that integrates Internet of Things (IoT) sensors with Machine Learning and Deep Learning techniques to improve prediction accuracy and support real-time monitoring. IoT sensors continuously collect environmental parameters, including temperature, humidity, soil moisture, and light intensity, while crop leaf images are acquired for visual analysis. The collected sensor data is processed using Machine Learning algorithms such as Random Forest, Support Vector Machine (SVM), and XGBoost to predict disease occurrence based on environmental conditions. Simultaneously, Deep Learning models, including EfficientNetB0 and Xception, are employed to classify crop diseases from leaf images by extracting high-level visual features. The fusion of sensor-based and image-based information enhances the robustness and reliability of disease prediction. Experimental results demonstrate that the proposed hybrid framework achieves superior prediction accuracy, faster inference, and better generalization compared with conventional approaches. Furthermore, the integration of IoT-enabled monitoring with intelligent analytics reduces manual inspection, supports timely disease management, and promotes sustainable agricultural practices. The proposed system provides an efficient decision-support solution for precision farming by improving crop health monitoring, optimizing resource utilization, and increasing agricultural productivity.