IoT-Enabled Smart Horticulture: Integrating Image Processing for Real-Time Environmental Monitoring and Disease Detection

Traditional horticulture continues to suffer from inefficiencies due to its dependence on manual monitoring, delayed identification of plant diseases, and excessive use of resources, which collectively reduce productivity and increase operational costs. To address these challenges, this research introduces an integrated framework that combines IoT-based multi-sensor monitoring with image processing supported by edge-level CNN-based disease detection and a cloud-enabled decision support system. Experimental validation demonstrates notable outcomes, including 94.2% disease detection accuracy, a 32% reduction in water usage, a 35% decrease in pesticide consumption, and a 22% improvement in crop yield under controlled greenhouse conditions. Owing to its scalability and practicality, the proposed system can be effectively deployed in greenhouses, polyhouses, and open-field horticultural environments to facilitate precision agriculture, optimise resource utilisation, and support sustainable crop management.