一种新的玫瑰扫描目标跟踪边界重心法

针对自动寻的导弹玫瑰扫描目标跟踪方法中由于扫描过程非线性引起的目标采样非线性,导致求取目标重心算法的运算量和计算误差大等问题,提出了一种基于格林公式的、新的求取目标重心的边界重心法,该方法仅使用目标边界上的采样点。通过对运算量和计算误差的理论分析,与现有的权重重心法相比,本方法可以有效解决采样不均匀的问题,同时运算量减少了90%,误差仅为权重重心法的下界。另外,本方法也可用于基于图像信息的凝视目标跟踪等类似算法中来提高跟踪准确性及运算速度。

A New Boundary Center of Gravity Method for Target Tracking in Rosette Scanning

The non-linearity of a rosette scanning process causes the non-linearity of target sampling which may account for the considerable calculation cost and large errors of the traditional methods.To solve this problem,this paper proposes a new approach based on Green’s theorem for computing center of gravity in rosette scanning.Only the samples on the boundary are used for computing the center of gravity of a target;therefore,the method can eliminate the disadvantage of non-linearity in a rosette pattern.Theoretical analysis and simulation demonstrate that the calculation error by this method is smaller than that of the weight gravity method and the cost is reduced by 90%.In addition,this method can also be used in image-based gaze tracking and other similar methods to improve accuracy and speed.