摘要
针对基于MeanShift的跟踪算法对目标尺度变化适应能力差并且不能对目标的旋转进行跟踪的缺点,提出了一种基于分块颜色直方图的MeanShift跟踪算法。新算法引入目标旋转和缩放矩阵使得算法能够适应目标的旋转和尺度变化,分块的颜色直方图包含了目标的空间信息,提高了跟踪算法的鲁棒性和适应能力。实验结果表明新算法能够同时对目标的尺度和旋转变化进行稳定的跟踪,改善了原算法的适应能力。
The tracking algorithm based on mean shift is incapable of adapting to the target rotation and size changing well. A new tracking algorithm based on block color histogram and mean shift was proposed. A target scale and rotation matrix was imported into the algorithm and made the algorithm be able to adapt to target rotation and scaling. The robustness and adaptability of the tracking algorithm was improved since the block color histogram included the spatial information of target. Experimental results show that the new algorithm is able to adapt to object rotation and sealing simultaneously and more practical than the original algorithm.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2009年第10期2936-2939,2955,共5页
Journal of System Simulation
基金
国家自然科学基金项目(60632050
60472060)
江苏省科技计划高技术研究项目(BG2005008)
