Smart Cities and AI: Real-Time Traffic Management and Energy Optimization with IoT Integration

This article delves into using AI in smart city systems, specifically at energy optimization and real-time traffic control. To alleviate traffic and increase mobility, AI-powered traffic management systems examine real-time data collected from cameras, sensors, and linked vehicles using computer vision and machine learning. Furthermore, public utilities and smart grids can benefit from improved resource allocation, predictive maintenance, and dynamic load balancing. Urban living standards, environmental impact, and economic savings have all been shown to improve significantly significantly in case studies of smart around the world. This study presents a new model for object identification in vehicles called a deep learning-based single-shot multi-box detector (SSD) to recommend an improved inception network that maximizes SSD and improves the small object feature extraction (FEA) of the shallow network. The second is to improve the fusion effect by modifying feature pyramid network. Then, deep feature fusion module enhances the deep network’s FEA. The research model incorporates improved version of sea horse optimization (MSHOA), which enhances classification accuracy, to fine-tune the parameters of the suggested model. Utilizing video footage captured by a closed-circuit television (CCTV) system on roadways in Tashkent, empirical findings demonstrate that the suggested model outperforms state-of-the-art approaches by around 98.1%.