【摘要】 提出一种基于特征代价函数的多目标跟踪算法。根据目标在相邻帧间运动具有连续性以及包围窗口、灰度变化不大的特点,改进了一种代价函数。在跟踪的匹配过程中,启动卡尔曼滤波和预测目标匹配搜索区域,同时使用目标链记录目标最新的运动状态和特征值,从而保证运动跟踪的连续性和计数的有效性。更多还原

【Abstract】 This paper proposes a method of multi-object tracking based on cost function of features.In light of the moving continuity and the small change of object’s features in frames, a kind of cost function is put forward for the tracking algorithm, which is applied in the tracking match of the image. At the same time, the Kalman filter is used to predicate the search areas of the matching objects, and moving state and feature value of the objects in current frame are recorded to ensure the continuity and effectiveness of the dynamic tracking by means of the object-chain.更多还原