关键词: 目标跟踪, GM-CPHD滤波, 初等对称函数, 时间复杂度, 执行效率

Abstract: In the problem of multi-target tracking, GM-CPHD filter can estimate recursively the targets’ information of the cardinality distribution and the intensity distribution simultaneously, and then the filtering results are more accuracy. However, one of its disadvantages is that the time complexity is high, especially in the case of a high noise rate, the filtering time is too long. The main reason is that the high computational complexity of the Elementary Symmetric Functions（ESF） in the update step of the algorithm. To solve this problem, this paper replaces the definition method with a recursive method to calculate ESF, and the efficiency of the GM-CPHD is improved. The experimental results show that the ESF calculated by the recursive method does not affect the accuracy of the filter, and can reduce its complexity.

Key words: target tracking, GM-CPHD filter, elementary symmetric function, time complexity, execution efficiency