AI-based Intelligent Traffic Management and Traffic Rule Violation Detection Using Computer Vision

This study investigates the use of artificial intelligence and computer vision for intelligent traffic management and automated traffic-rule violation detection. The proposed framework combines video-based vehicle detection, traffic-density estimation, rule-based violation detection, automatic number plate recognition (ANPR), optical character recognition (OCR), centralized violation logging, and administrative monitoring. A mixed-method approach was used, combining a pilot survey of 12 participants with an engineering-oriented system architecture. The pilot findings indicate support for density-based adaptive traffic signals and a preference for human verification before automated enforcement. The study also identifies challenges related to occlusion, adverse environmental conditions, real-time processing, scalability, privacy, and evaluation consistency. The proposed framework provides a basis for future implementation and empirical validation of AI-assisted traffic management systems.