一种复杂场景下的运动目标跟踪算法

提出了一种基于跟踪窗口自适应和抗遮挡的目标跟踪算法。采用Mean Shift算法确定当前帧的目标位置,最优选取核函数带宽,使跟踪窗口能够根据目标尺寸大小作出自适应调整。利用Bhattacharyya系数作为遮挡的判断依据,当目标遮挡时引入卡尔曼滤波器估计目标的运动信息,进行后续状态预测。实验表明,该算法能有效跟踪复杂场景下的运动目标。

A moving target tracking algorithm under complicated scenes

A target tracking algorithm which is based on adaptive tracking window and anti-occlusion is presented in this paper.Mean Shift algorithm is used to determine the target position in the current frame,and the bandwidth of the kernel function is optimally selected,which can make the tracking window adaptively adjust according to the target size.Bhattacharyya coefficient is used as a judgement criterion for occlusion.Kalman filter is introduced to estimate the moving information of target and make the prediction of the follow-up state when occlusion happens.Experimental results show that this algorithm can effectively track the moving target under complicated scenes.