AI-DRIVEN SURVEILLANCE & PRECISION CONTROL OF CURVULARIA LUNATA A RESNET-BASED APPROACH FOR SUSTAINABLE MUSKMELON HEALTH

Abstract

Muskmelon (Cucumis melo), a widely cultivated fruit crop, is highly susceptible to fungal pathogens that affect its leaves, stems, roots, flowers, seeds, and fruits, leading to considerable economic losses. Among these pathogens, Curvularia lunata is a primary threat, responsible for leaf blight, fruit rot, root decay, seedling damping-off, and flower necrosis, ultimately reducing crop viability and market quality. Traditional disease detection methods rely on visual inspections and conventional diagnostics, which are time-consuming, prone to errors, and often delay critical interventions. To overcome these limitations, this study integrates AI-powered surveillance using the ResNet deep learning algorithm, which processes high-resolution imagery from UAVs, spectral sensors, and in-field cameras to classify disease severity with high accuracy and precision at different plant growth stages.By leveraging pre-defined datasets and real-time monitoring, the ResNet model facilitates early detection of disease symptoms across multiple plant parts, enabling data-driven decision-making for targeted interventions. The model's performance was validated using key statistical methods, including ANOVA, paired t-tests, and regression analysis, which demonstrated a significant correlation between AI-assisted disease detection and optimized disease management strategies. Implementing AI-driven precision fungicide application schedules significantly reduced disease progression and enhanced overall plant health. The study underscores the transformative role of AI in sustainable disease management, ensuring reduced fungicide dependency, optimized resource allocation, and increased muskmelon yield. This approach also establishes a foundation for broader applications in the Cucurbitaceae family, advancing precision agriculture and eco-friendly crop protection strategies.