摘要
提出一种非常有效且具有良好鲁棒性的人体目标跟踪算法。由于传统的卡尔曼滤波不能很好地解决非线性、非高斯问题的跟踪,为此提出了一种新型的粒子滤波器跟踪算法。该算法采用加权的粒子集模型表示状态的分布,用迭代运算跟踪状态的变化,从而有效地解决了数据处理的量大和模型出现高维的问题。实验结果证明,该算法对固定摄像机单一背景下人体目标跟踪是快速且有效的。该算法可广泛应用于航空器位置的跟踪、噪声环境通信信号的估计、人体或车辆的跟踪。
A very efficient and robust human-body object tracking algorithm is presented. Because of Kalman filter is inadequate in solving the problems of non-linear and non-Gaussian model, a new particle filter tracking algorithm is proposed. This method represents the distribution of the states by weighted particle model and tracks the of states' change with iterative calculation, in this way it resolves the problems of huge data processing and high-dimensional model effectively. Experimental results show that the presented method is fast and effective to the tracking of human-body object in uniform background condition of fixed camera. What's more, this algorithm can be used in tracking of aircraft positions, estimating communication signals in noisy environment, and tracking of people or cars in surveillance videos.
出处
《计算机应用与软件》
CSCD
北大核心
2008年第12期247-248,251,共3页
Computer Applications and Software
关键词
粒子滤波器
人体目标跟踪
贝叶斯算法
鲁棒性
Particle filter Human-body object tracking Bayesian algorithm Robust