Real-time traffic data can be analyzed by AI algorithms, and traffic signals can be adjusted accordingly, reducing travel time and fuel consumption. Despite the existing solution, the traffic jams are hard to detect and pinpoint, and determining their causes remains challenging. This research provides a unique solution by forecasting the jams using sensor data, leading to decision-making for the officials to regulate the traffic, along with the IoT sensor data, the research also embeds the drone and CCTV data to identify the areas with congestion and provide live analysis of the traffic situation. In this research, we have created an integrated system that uniquely combines IoT, drone, and CCTV AI models, providing a one-stop solution for officials, both cloud managers and hardware managers. This AI solution has been trained and tested on diverse datasets and scenarios to classify up to twelve classes of objects, achieving 87 % accuracy by combining machine learning and deep learning models. Therefore, this research has proved to be one of the advanced transportation solutions that can lead to real-time congestion analysis, fuel efficiency, and operational cost reduction as we provide the drone simulation. Furthermore, it opens up new opportunities for investment in the drone industry, paving the way for improved traffic monitoring and management.
Smart City AI Traffic Cloud - City Traffic Analysis and Monitoring Using AI Models
2025-05-05
3901986 byte
Conference paper
Electronic Resource
English