Adjustment of Surveillance Video Systems by a Performance Evaluation Function

摘要

This paper proposes an evaluation metric for assessing the performance of a video tracking system. This metric is applied to adjust the parameters that regulates the video tracking system in order to improve the system perfomance. Thus, the automated optimization method is based on evolutionary computation techniques. The illustration of the process is carried out using three very different video sequences in which the evaluation function assesses trajectories of airplanes, cars or baggage-trucks in an airport surveillance application.