Moving target detection in the binocular vision system

The video surveillance system has been widely used in the field of security monitoring. The binocular vision system could provide more information of monitored targets and has unique advantages in analysis of suspicious targets and understanding of images. Currently, the research and application of the binocular vision system has been widely concerned by domestic and foreign researchers. Moving target detection is developing and innovating as the primary problem in the most visual system. In this paper, three classic algorithms of moving target detection have been introduced: background difference method, frame difference method and optical flow method. After compassion and analysis of the three algorithms through the experiment, the study concludes that the background difference method based on the mixed Gaussian model has some advantages in binocular vision system.