Incheon National University, Inch'on-jikhalsi, Republic of Korea
Abstract: The accelerating spread of vector-borne diseases, driven by climatic and ecological shifts, highlights the need for continuous, intelligent, and scalable mosquito surveillance. Conventional monitoring programs, often based on manual sampling and delayed reporting, remain limited in temporal resolution and responsiveness. To achieve effective control, surveillance must evolve toward a data-driven and proactive paradigm aligned with the principles of Integrated Pest Management (IPM). We present an AI-IoT integrated surveillance framework that enables real-time collection, transmission, and analysis of entomological and environmental data. Field-deployed smart traps equipped with optical and environmental sensors autonomously acquire high-frequency data on mosquito activity, species composition, and habitat conditions. Cloud-based analytics and predictive modeling transform these continuous data streams into evidence-based insights for spatially and temporally targeted interventions. The approach enhances accuracy, efficiency, and scalability while reducing labor intensity and ecological impact. Beyond immediate operational gains, the accumulation of large-scale longitudinal datasets establishes the foundation for predictive and service-oriented applications. These may include trend forecasting, adaptive risk modeling, and interactive advisory systems that support real-time decision-making for public health and pest management authorities. Looking forward, emerging technologies such as Vision-Language and multimodal AI models could enable context-aware, query-based analysis, allowing future systems to interpret ecological patterns and recommend optimized control strategies in natural language. In summary, this work illustrates how the convergence of AI, IoT, and IPM can transform mosquito surveillance from reactive observation to predictive, data-centric ecosystem management, advancing the long-term goals of sustainable and intelligent vector control.