Optimization of Traffic Congestion in Smart Cities Using Residual Convolutional Neural Network

With increasing density of vehicle in smart cities, the traffic gets worsen day by day, therefore it is necessary to optimize the traffic signals for smooth flow of traffic. In this paper, we develop a real-time solution on traffic signal control for the reduction of traffic congestion. The study develops a ResNet approach using Internet of Things (IoT) that controls the traffic congestion in smaller congestion area. The real-time analysis generates the traffic simulation environment in a simulator using the real time data i.e., finding number of vehicles getting congested from the images captured via IoT image acquisition module. The simulation generation using ResNet generates the control signal to real-time environment to quickly clear the congestion in that area. The experimental results with the support of simulator shows that the proposed ResNet is efficient to control the traffic congestion in smart cities.