计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (12): 50-52.

• 学术探讨 • 上一篇    下一篇

基于均值漂移与卡尔曼滤波的目标跟踪算法

常发亮 刘雪 王华杰   

  1. 山东建筑工程学院 计算机系 山东大学控制科学与工程学院 上海市水务信息中心
  • 收稿日期:2006-05-22 修回日期:1900-01-01 出版日期:2007-04-20 发布日期:2007-04-20
  • 通讯作者: 刘雪

Target tracking algorithm based on meanshift and Kalman filter

Xue Liu   

  • Received:2006-05-22 Revised:1900-01-01 Online:2007-04-20 Published:2007-04-20
  • Contact: Xue Liu

摘要: 均值漂移算法在目标跟踪过程中没有利用目标的运动方向和速度信息,在目标受到干扰时容易跟踪失败,而Kalman滤波能够较为准确地预测目标的速度和位置。因此本文提出一种结合均值漂移与Kalman滤波的跟踪算法,使用Kalman滤波对目标运动速度和空间位置进行预测。根据干扰的不同情况,使用不同的比例因子将两算法的跟踪结果线性加权得到目标的最终位置。实验结果表明该算法是可行有效的。

Abstract: Meanshift algorithm doesn’t use the target’s motion direction and speed information in process of target tracking. When affected by disturbance it easily fails to track the target. Kalman filtering can predict the position and velocity of the target exactly. An algorithm combined Kalman filtering with meanshift algorithm is proposed in this paper. Kalman filtering is used to predict the position and velocity of the target. According to different disturbance circumstances, the two algorithms tracking results are done with liner weight method by using different scale factors to get the final position of the target. Experimental results show the good performances of the proposed algorithm.