说明：  为了提高经典的Mean Shift算法在复杂场景中的跟踪性能,提出了一种基于角点的目标表示方法。首 先,利用Harris角点检测算法提取表示目标主要特征的角点 其次,基于提取的角点,建立目标模型,将其嵌入 Mean Shift算法进行跟踪。该方法仅用少量的关键点表示目标,能够自动去除目标和背景中的次要特征,有效地 抑制背景成分对目标定位的影响,从而改进Mean Shift目标跟踪算法的性能。
(To imp rove the performance of the classicalMean Shift tracking algorithm in the comp lex scene, developed a target rep resentation method based on corner points. First, detected the corner pointswhich contain themain features of the object by using Harris corner detector. Then used the obtained corner points to calculate the target model, which was furthermore em2 bedded into theMean Shift tracking algorithm.)