Detection and Parameters Estimation of Moving Objects via Video Surveillance

Abstract

The article proposes an algorithm for processing video information obtained from traffic monitoring in order to automatically detect moving objects and evaluate some of their parameters. The algorithm automatically detects passing vehicles, and evaluates their dimensions, speed and direction of movement. The algorithm does not require large computational resources and can work in real time. It is proved on the real video traffic records and the results of the examination becomes very close to the real ones. It is applicable in smart traffic management systems with CCTV. In the future, the algorithm will be extended with capabilities for cars recognition and their license plates identifying.