IoT-Integrated Leaf Analysis for Early Plant Disease Detection in Agriculture Using Hybrid Ensemble Models

Crop diseases threaten food security and agricultural production. For prompt interventions, disease management, and loss reduction, crop disease outbreaks must be detected early. Precision Agriculture (PA) promotes sustainable agriculture by developing predictive plant disease prevention technology using ICTs. This paper presents a multi-step illness detection system using contemporary sensors and images. System architecture includes preprocessing, feature extraction, model training, and disease detection. Crop conditions are monitored using temperature, colour, and humidity sensors. High-resolution (HR) picture preprocessing improves raw images for predictive analysis. Leaf pictures provide geometric characteristics including area, perimeter, and rectangularity to improve model correctness. Model Driven ERL trained a plant disease prediction model. Combining sensor data with machine learning algorithms in PA yielded 89.47% plant disease prediction accuracy. A scalable technique to early disease identification in agriculture makes this technology potential for proactive plant disease control. This model's precision makes it suitable for real-world PA systems, reducing disease-related crop losses and ensuring agricultural output sustainability.