基于改进Mean Shift算法的细胞追踪方法研究

针对经典Mean Shift算法不能有效追踪快速移动细胞的缺陷,提出了利用Mean Shift和卡尔曼滤波器相结合的方法快速移动细胞进行追踪。算法以卡尔曼滤波器预测出细胞的位置作为Mean Shift算法的初始位置,然后再利用Mean Shift算法追踪得到的细胞位置作为下一帧的卡尔曼滤波器的输入参数。实验结果表明,对于细胞图像的追踪,该方法较经典Mean Shift算法有着更高的准确率。

Tracking method for cells based on improved Mean-Shift algorithm

To improve the limitation of classical Mean Shift algorithm for tracking fast moving cells.This paper proposed a tracking algorithm which has combined Mean Shift and Kalman filter to track fast moving cells.The starting position of Mean Shift is predicted by Kalman filter,then Mean Shift track the cells position which is used to update Kalman filter parameters.The experimental results show that the approach has improved tracking accuracy compared with classical Mean Shift algorithms.