摘要
动态背景下的运动目标跟踪算法一直是计算机视觉领域的研究热点。针对动态背景下的运动目标跟踪,提出一种基于点对矢量场的运动目标跟踪算法。首先,提取出运动目标身上的点对特征点。然后根据灰度、梯度方向差和点对距离的一致性对运动目标的特征点进行匹配,并根据匹配点对构建点对矢量场。最后,根据点对矢量场实现对运动目标的跟踪。实验结果表明,该算法对动态场景下的运动目标跟踪具有一定的有效性。
Moving object tracking algorithm based on dynamic background has been a hot research topic in the field of computer vision. For moving object tracking in dynamic background, presents a new algorithm of moving object tracking based on the vector field of point pair. Firstly, extracts the point pair of the moving objects. Then, matches the feature points of moving objects according to the consistency of gray scale, gradient direction difference and distance of the point pair. And the vector field is constructed according to the matching points. Fi- nally, tracking of moving objects based on the vector field of point pair. Experimental results show that the proposed algorithm is effective for the moving target tracking in dynamic scenes.
出处
《现代计算机》
2016年第20期11-14,共4页
Modern Computer
关键词
目标跟踪
点对
矢量场
动态背景
Target Tracking
Point Pair
Vector Field
Dynamic Background
