中文摘要：

研究视频图像运动目标跟踪识别问题时，目标受纹理、光照变化、遮挡及与目标差异较大的矩形特征不断积累，易导致跟踪漂移或丢失。大多数跟踪系统通过牺牲跟踪的实时性来提高跟踪的准确性。为了解决上述问题，引入两个随机投影矩阵提取两类互补的纹理特征和灰度均值特征，利用随机投影矩阵提取的特征构造候选特征池。若候选目标某一区域受遮挡时．采用未被遮挡区域的矩形特征构造特征池。并选取反应目标特点的矩形特征，去除与目标特征差异较大的矩形特征构造分类器；然后计算互补特征对样本的分类权值，选权值较大的特征作为寻找目标的下一帧位置；最后根据分类器分类分数最大所对应的候选区域确定目标的下一帧位置。仿真结果表明：改进方法具有更好的跟踪精度，且计算时间少，对目标纹理、光照变化和遮挡具有更强的鲁棒性。

英文摘要：

In order to improve the accuracy of moving target tracking in video image, two random projection matrices are introduced to extract two kinds of complementary texture features and gray mean feature, and the projection matrices are used to extract the rectangle feature to construct a feature pool. If the candidate object is occluded, the feature pool is constructed by the rectangle feature of the occluded region. The rectangle feature is used to represent the characteristics of target in the feature pool, and the rectangular features with greater difference from the target characteristics are removed. Then the complementary characteristics of the classification of sample weights are calculated, the selected large weight features are used to find the target in next frame. Finally, the classifier is taken to process candidate samples by Bayes classification and response results to the classifier are taken as tracking results. The simulation results show that the proposed algorithm has small computation time and can accurately capture the tracking target in target textures, lighting change and complex background or occlusion.