摘要
检测前跟踪方法在计算似然函数时需要计算每一个分辨单元,若分辨单元数目较多,将使得计算量较大而影响检测跟踪的实时性。该文提出了一种快速实现方法,在计算似然函数时仅选取受到目标能量扩散影响较大的分辨单元,从而降低计算量。最后采用粒子滤波检测前跟踪方法进行了快速实现仿真,并分析了部分参数对弱目标检测跟踪的性能影响。
Track before detect(TBD) computes every resolution cell to get likelihood function and thus costs amount of time when there are large number of cells. A fast implementation of track before detect is proposed to solve those problems. Only cells that primarily affected by the target are considered to compute the likelihood function in order to save calculation. Simulations of the fast implementation of the particle-based TBD are taken in order to analyze the performance effect of parameters.
出处
《杭州电子科技大学学报(自然科学版)》
2013年第3期61-64,共4页
Journal of Hangzhou Dianzi University:Natural Sciences
关键词
弱目标
检测前跟踪
粒子滤波
局部似然
weak target
track before detect
particle filter
local likelihood