IOT - ENABLED SMART DUSTBIN NETWORK WITH AI BASED GARBAGE REPORT VERIFICATION SYSTEM

This paper presents the Smart Garbage Management System (SGMS), a three-component platform that addresses
inefficiencies of traditional waste collection in dense urban zones. The system integrates an IoT smart bin equipped
with an fill-level sensor, and LED status indicators; a browser-based management centre dashboard that displays live
bin fill levels across multiple locations in specific zone on an OpenStreetMap base map and auto-generates demand-
driven collection schedules; and a mobile-first public reporting application through which citizens submit AI-verified
photographic evidence of illegal dumping. AI Vision API performs automated image analysis and rejects non-genuine
submissions before they reach the operations team. All components communicate via MQTT over the HiveMQ public
broker. The complete system is deployed as a web application at sgms26.web.app on Firebase Hosting. Evaluation
demonstrates sensor accuracy within 1.6% mean absolute error, end-to-end MQTT latency below 800 ms at the 95th
percentile, and AI verification accuracy of 93.3% across 30 test images.