ACCOUNTING DATA UNCERTAINITY IN MODELING EVENTS WITH PETRI NETS

Abstract

The analysis of the problem of recognition events in the video was conducted, identified the main tasks set in the development of systems of recognition events. The use of Petri nets as a model for recognition events was examined. The  probabilistic model based on the Petri net in the form of the recursive Bayesian filter was proposed.. Testing conducted on a set of videos and shows the effectiveness of the proposed model under uncertainty of input data using real tracking algorithms.