AI-Enhanced IoT System for Efficient Traffic Management: Leveraging Black Data in Smart Cities

This paper introduces an artificial intelligence (AI) reinforced Internet of Things (IoT) infrastructure specifically developed to tackle the pressing problem of unused black data in urban traffic management. The framework combines Convolutional Neural Networks (CNNs) and Support Vector Machines (SVMs) with IoT -enabled speed track cameras to transform large volumes of previously disregarded traffic data into practical and significant insights. The architecture of the system effectively screens and analyzes data at the edge before sending pertinent information to cloud storage, significantly reducing the strain on cloud infrastructure. The simulation results highlight the efficacy of the framework by attaining a 95% accuracy rate in identifying traffic infractions and decreasing processing time to an average of 2 seconds per data packet. In addition, the system effectively employed 80% of the black data, resulting in enhanced and informed traffic management decisions and increased urban mobility. This study presents an innovative method for redistributing underused data, hence improving the sustainability and effectiveness of traffic systems. The results indicate that the framework is highly suitable for operational implementation in intelligent urban settings and has the potential to be expanded to larger metropolitan settings, paving the way for advanced, data-driven traffic solutions in smart cities.