基于DM6437的目标跟踪系统设计与实现

Design and Realization of Target Tracking System Based on DM6437

将卡尔曼滤波器和均值漂移算法相结合,实现运动目标的实时跟踪.首先采用混合高斯背景模型的方法来进行运动目标的检测,确定运动目标的大小和位置,然后提取运动目标的颜色模型并开始跟踪.在跟踪过程中,先用卡尔曼滤波器对目标在下一帧中的位置进行预测,再用均值漂移法快速迭代得出目标的实际位置,最后更新目标的颜色模型,并开始下一轮跟踪过程.采用高性价比的DM6437芯片设计了运动目标跟踪系统,并对跟踪算法进行了优化.实验结果表明,该系统具有较好的实时性和准确性.

To achieve real-time tracking of moving targets, Kalman filter and mean shift algorithm were used. Firstly , the mixed Gaussian background model was used to detect moving target, de-termine the size and position of moving targets, and then moving object color model was extracted and the system began its tracking. In the tracking process, Kalman filter algorithm was used to predict the target position of the next frame first, then mean shift algorithm was used to find out the targets actual position, target color model was updated according to results of Kalman filter and mean shift, after which the next tracking process was began. The moving target tracking system based on DM6437 of high cost performance was designed, and the tracking algorithm was optimized. Experimental results show that the system has good real-time performance and accuracy.