VEHICLE COUNTING FOR TRAFFIC MANAGEMENT SYSTEM USING YOLO AND CORRELATION FILTER

The effective control of traffic flow is a significant difficulty that urban regions all over the world must overcome. Vehicle counting is crucial to traffic management systems because it gives information that is necessary for designing infrastructure, analysing traffic flow, and reducing congestion. Urban regions are increasingly plagued by traffic congestion, which increases travel time, fuel use, and air pollution. In order to mitigate these problems, effective traffic management systems are essential. Vehicle counting is a crucial part of these systems since it offers useful data for planning infrastructure, traffic light optimisation, and traffic flow analysis. Deep learning and computer vision approaches have demonstrated promising results in object tracking and detection applications in recent years. In order to recognise and track vehicles accurately and effectively, this research provides a vehicle counting system that combines the YOLO object recognition technique with CF tracking. The You Only Look Once (YOLO) object identification algorithm and Correlation Filter (CF) tracking are used in this research study to present a unique method for vehicle counts. We seek to achieve accurate and real-time vehicle counting for efficient traffic management by integrating the advantages of both approaches.