Several algorithms of remote objects capture and tracking in a video stream, such as Lucas - Kanade, Motion Templates, Mean-Shift, CamShift, Viola - Jones were considered. A comparative analysis of algorithms, identified their strengths and weaknesses was conducted. The efficiency of each algorithm applied to the problem tracking remote object presents the results of their work were presented. By analyzing the probability of error types I and II with each of the discussed algorithms and their characteristics in the processing stream, two of the most suitable algorithm for capture and tracking were allocated. We propose an integrated algorithm that uses the capabilities of the algorithm Lucas - Kanade and the algorithm based on Haar cascades to capture and tracking the remote object.