Review of Computer Vision-Based Vehicle Tracking Applications and Techniques
Video processing techniques provide a reliable, efficient, and scalable approach to different traffic applications
due to their low deployment and maintenance cost as well as the capacity of wide area coverage. In computer vision,
motion tracking refers to the task of estimating the location of a single or multiple moving object over the time in a
sequence of images captured by a video camera. Vehicle tracking, in general, is a challenging task, due to the change of
vehicle motion, appearance, occlusions, and camera motion. Hence, the aim of this paper is to review of the existing range
of video processing applications for traffic monitoring and safety, which include: traffic density detection and vehicle
counting, automated recognition of license plate characters, and video analysis for anomaly and on-road incident detection.
Moreover, to review the current state of the art tracking methods, including vehicle tracking and traffic surveillance
applications based on Kalman filter-based tracking. In this paper, we categorise vehicle tracking methods into three
categories: Feature-based method, Kernel-based method, and Silhouette-Based method.
Keywords - Tracking, Kalman Filter, Mean Shift, Vehicle Tracking,