视频中快速运动目标的自适应模型跟踪算法

Adaptive Model Tracking Algorithm for Fast-Moving Targets in Video

针对现有视频跟踪算法因目标运动机动性强或非对称刚体目标快速变形等情况导致的目标丢失问题,提出了一种基于相关滤波自适应模型与平均峰值相关能量(APCE)重检机制的视频跟踪算法。自适应模型跟踪算法可根据目标区域的清晰度实时调整模型,有效保证目标跟踪模型的准确性。实验结果表明,将自适应模型跟踪算法融入判别式尺度空间跟踪(DSST)模型中,可增强模型对机动性强或快速变形物体的跟踪效果,在保证跟踪速度的同时,使原始DSST模型的平均精度提高了18.3个百分点,成功率提高了15.2个百分点。此外,将自适应模型跟踪算法与APCE重检机制相结合,可保证跟踪算法的稳定性。

To address the problem of target loss facing existing video tracking algorithms due to high mobility of targets or rapid deformation of asymmetric rigid targets,this paper proposes a video tracking algorithm based on the correlation filtering adaptive model and the redetection mechanism for average peak-to-correlation energy(APCE).The adaptive model tracking algorithm can adjust the model in real time according to the clarity of the target area,thereby effectively ensuring the accuracy of the target tracking model.Experimental results show that integrating the adaptive model tracking algorithm into the discriminative scale space tracking(DSST)model can enhance the tracking effect of the model on highly mobile or rapidly deforming objects.While guaranteeing tracking speed,the integration also raises the average accuracy of the original DSST model by 18.3 percentage points and the success rate by 15.2 percentage points.In addition,combining the adaptive tracking algorithm with the APCE redetection mechanism can ensure the stability of the algorithm.