| Summary: | The article deals with the problem of correct detection of moving objects when they are monitored by the monitoring system that operates in the visible range of the electromagnetic radiation spectrum. Based on the analysis of existing methods, for detection of moving objects there is proposed an algorithm that is adaptive to destabilizing factors. This algorithm makes it possible to increase the accuracy of moving objects detection. The algorithm takes into account the presence of noise and its heterogeneity, both in space and time, and also removes the influence of moving shadows. The correctness of the algorithms described in this article is confirmed by their software implementation and modelling. In the process of modelling, the accuracy of object detection, proposed by the algorithm under different observation conditions, and the motion parameters of the objects were estimated.
Keywords: video monitoring, detection of moving objects, detection of moving shadows, background subtraction, interframe difference
|
|---|
