Method of dynamic image`s identification in videosurveillance systems of critical infrastructure

Abstract

It is established that the efficiency of critical infrastructure depends on the high level of video surveillance systems informatization. At the same time, continuous improvement of technical equipment of these systems requires the creation of new and improvement of existing methods of detection of scan movements. One of the promising approaches is the development of motion detection methods on a static image background in intelligent video surveillance systems. Timely detection of unauthorized access to the territory of critical infrastructure is a necessary component of ensuring high stability and reliability of their functioning, especially in the face of the threat of terrorist acts and necessitates the development of methods of detection of movements and unauthorized access to the sensitive territory. To this end, the article proposes a method of identification of dynamic images, which is based on the algorithm of background subtraction. The results of verification of the method for different types of correlation allocation, their comparative characteristics are given, advantages and disadvantages are analyzed. It is shown that the application of the proposed method makes it possible to quickly identify the facts of penetration of critical infrastructure objects into the territory of the regime and to prevent unauthorized access to them in the absence of means of motion detection.