Published Paper Details:

In modern urban environments, the enforcement of traffic regulations remains a significant challenge due to increasing vehicle density and limited enforcement personnel. This project presents a scalable, AI-driven traffic violation monitoring system that automates the detection, documentation, and notification of traffic offenses such as helmet non-compliance and stop-line crossing. The system leverages a YOLO-based object detection model for real-time analysis of CCTV feeds, coupled with image processing techniques for accurate violation identification. An integrated OCR-based ANPR module extracts license plate numbers from captured frames, enabling automated challan generation and offender identification. The backend, powered by a Flask server, manages violation records, initiates Stripe-based payment processing, and renders geo-visualizations through Google Maps API. To enhance field responsiveness, a GPS-enabled Android application tracks traffic officers and dispatches Firebase Cloud Messaging (FCM) alerts to nearby personnel upon violation detection. Additional features include WhatsApp notifications and Google Text-to-Speech (TTS) announcements to expand alert reach and improve engagement. Tested in simulated environments mimicking urban intersections, the system demonstrated high accuracy in de- tection and efficient communication workflows. With support for automated alerts, digital payments, and real-time enforcement coordination, the proposed architecture offers a comprehensive and responsive solution for modern traffic law enforcement systems.