Mean－shift算法是一种非参数密度估计算法，可以实现快速的最优匹配，在目标的实时跟踪领域起着非常重要的作用。为了有效的将Mean－shift算法应用到灰度图像中，采用了以方向直方图建立目标模型的策略，提出了在灰度图像中以Mean－shift为核心的目标跟踪算法。实验结果表明，该算法具有不受光照条件影响的优点，在低对比度的情况下仍能实现稳定、实时的跟踪目标。

The Mean-shift algorithm is a nonparametric density estimation method. The fast and optimal mode matching can be achieved by this method. This algorithm plays a very important roles in real-time object tracking field. In order to apply Mean-shift algorithm on gray-level image,a strategy that using orientation histogram to represent target is adopted. A new algorithm based on Mean-shift is established. Experiment results show that this algorithm can adapt the change of illumination. Though the contrast of image is low,the object can be tracked robust and real-time using the new algorithm.