Development of tracking and segmentation algorithm of partially occluded moving objects

Monitoring, tracking and identification of moving objects is one of the most significant task of modern videosurveillance systems that uses machine vision methods. Today such systems are suitable for many applications due to their operating flexibility and operating convenience. Using machine vision methods, modern videosystems afford an opportunity to solve such problems as: identification of incorrectly parked vehicles, measurement a traffic parameters etc. A key task that arises during monitoring and tracking of moving objects is correctness of their segmentation. In this paper we propose an algorithm of segmentation and tracking of moving objects that partially occluded in videostreams.