Research on majorization strategies of vision system of RoboCup

In this paper, the algorithms of object tracking and identifying of vision system of RoboCup are researched, and the majorization strategies of improving the quality and rate of identifying greatly are made. These are sufficient to achieve a precise and fast system and to satisfy the need of real-time vision system. The error acceptance to error identifying is also mentioned in this paper, which enhances the correctness of the system. It helps the team of RoboCup of Shanghai University performance well in "Master" China Robot Competition in 2003.