A distance weight object tracking method based on combining mean shift and GM(1,1)

作者：Shi Meifeng^*; He Zhongshi; Liu Xin; Huang Meiyan

来源：International Journal of Digital Content Technology and Its Applications, 2012, 6(1): 318-325.

DOI：10.4156/jdcta.vol6.issue1.38

• Background filtering
• Complex background
• Data forecast
• Distance weight
• GM (1 , 1)
• Gray Model
• Mean shift
• Object Tracking
• Real time tracking
• Tracking accuracy

In the process of object tracking, classic Mean shift can not solve the following two key problems: Object tracking under complex background, object obstructed long time during tracking. GM(1,1) is applied in this paper to reduce the impact of complex background pixels by background filtering and to track the object by data forecast even when the long time obstructing. Distance weight is introduced to combine Mean shift and GM(1,1) to increase tracking accuracy. Thus, a novel distance weight object tracking method with combining Mean Shift and GM(1,1) (DWMG for short) is proposed in the paper. Finally, experiments results show that the proposed algorithm can improve the robustness of object tracking and ensure real-time tracking.

• 出版日期2012