基于DSP的多目标跟踪系统设计与实现

Design and Implementation of DSP-based Multi-target Tracking System

该文提出了一种基于卡尔曼滤波的多信息融合匹配的多目标快速跟踪算法,并在TI公司开发的DaVinci系列TMS320DM6446实现了多目标跟踪系统。该算法由卡尔曼滤波器预测和修正目标质心轨迹,然后融合质心和外接矩形框面积等多个信息对其进行匹配。实验结果表明,该系统具有良好的跟踪效果,有效解决目标遮挡后分离的跟踪情况,具有计算量小、存储量低、实时性高的优点。经过优化后,跟踪一帧图像时间约为50.2ms,实现实时跟踪。

This paper proposes a multi-information fusion matching multi-target tracking algorithm,which is based on Kalman filter,and achieves the system on the hardware platform DM6446.The algorithm firstly predicts and corrects the target＇s centroid trajectory by the Kalman filter,then finds the matching one with the centroid and the area of extern rectangle.Experimental results show that the system has a well tracking effect,and can track the correct targets after several targets occurred shelter.The system has the advantages of small amount of calculation,low storage capacity and high real-time.After optimization,it takes 50.2ms to tracking a frame of image,and achieves real-time tracking.