：该文把局部三值模式(Local Ternary Patterns, LTP)纹理特征引入Mean Shift 跟踪算法，提出了基于多

特征的Mean Shift 人脸跟踪算法以解决Mean shift 跟踪算法的鲁棒性问题。通过对LTP 纹理特征的分析、研究，

提出了一个LTP 关键纹理模型，既增强了目标的关键纹理信息，又简化了LTP 纹理模型。在此基础上，提出一

种基于LTP 关键纹理特征和肤色特征的Mean Shift 人脸跟踪算法，有效地解决了Mean Shift 算法的鲁棒性问题。

为进一步提高对快速运动目标的跟踪速度和跟踪性能，该文引入了卡尔曼滤波器对目标进行预测。实验结果表明，

该文的算法在目标定位的准确性和跟踪性能上比Mean Shift 算法均有明显的提高。-: In this paper, the texture characteristics of the local ternary patterns Local Ternary Patterns (LTP) Mean Shift tracking algorithm proposed Mean Shift face tracking algorithm based on multiple features in order to solve the Mean shift tracking algorithm robustness. Texture characteristics of LTP analysis, research, a key LTP texture model, not only enhanced the key goal of texture information, but also simplifies the the LTP texture model. On this basis, based on the LTP key texture features and color characteristics Mean Shift face tracking algorithm, effectively solved the robustness of the Mean Shift algorithm. To further enhance the fast-moving target tracking speed and tracking performance, this paper introduces the Kalman filter to predict the target. The experimental results show that the algorithm of the text in the target positioning accuracy and tracking performance than the Mean Shift algorithm to significantly improve.