Research and implementation of a distributed track marking system for mass video data

In order to achieve maximum efficiency and benefit through adjusting the places of business-related items, and dig out the frequent infested area that the customers appear and the goods they often purchased in the supermarket, We developed an interactive track tagging system. After gaining the target height information through the multi-camera, we can track the customers and label its trajectory. The experimental results show that the system is not only able to accurately track and mark the target trajectory in complex background, but also can significantly improve work efficiency at the same time. Because the distributed characteristics of the system make it possible to quickly deal with a large number of video data.