Design and implementation of an IoT-OL trap for community-based dengue early warning system

Surveillance of mosquito populations is central to successful vector intervention and disease control. Conventional surveillance techniques of manual egg counts and field-based assessments are cumbersome, slow, and hard to scale for real-time monitoring applications. To overcome these limitations, the paper introduces the design, implementation, and deployment of an Internet of Things (IoT)-Based Orvicidal-Larvidal (OL) trap and Community Dengue Early Warning System (C-DEWS) to facilitate the automated detection, enumeration, and monitoring of Aedes aegypti eggs in Cauayan City, Isabela. A Raspberry Pi equipped with micro-camera lenses captures the images and in determining the object of interest. The Convolution Neural Network (CNN) associated with the device achieves an outstanding 99.5% success rate for detecting and counting Aedes aegypti eggs, ensuring a reliable system. In addition, all key indicators, such as environmental factors, sanitary practices, Aedes aegypti egg distribution per location, and integration of dengue cases from the City Health Office (CHO) determine and monitor the dengue outbreak. This work shows how smart mosquito surveillance systems emerge from integrating data and multiple networks with cloud management frameworks. published by the Journal of Biodiversity and Environmental Sciences | JBES