摘要
基于动态规划的多目标检测跟踪算法涉及计算量爆炸和邻近目标难分辨的问题.针对实际应用中目标个数未知和环境中强弱目标并存限制的情况,提出一种基于动态规划的多目标复合检测与跟踪算法.首先利用传统恒虚警率(CFAR)算法进行一次检测,实现对强目标的检测和剔除;然后采用基于动态规划检测前跟踪算法对一次检测后的数据进行分区动态规划,实现对微弱目标的检测与跟踪,从而实现复杂环境下密集目标的有效探测.通过仿真和实测数据对本文算法性能进行了验证,结果表明该算法的高效性和稳健性.
The multi-target detection and tracking algorithm based on dynamic programming(DP)involves the problem of computational explosion and difficulty in distinguishing adjacent targets.Aiming at the unknown number of targets in practical application and the limitation of strong and weak targets co-existing in environment,this paper proposes a multi-target composite detection and tracking algorithm based on dynamic programming.Firstly,the traditional constant false alarm rate(CFAR)algorithm is used to perform the initial detection so as to identify and eliminate the strong targets.Then,the DP-based track-before-detect(TBD)algorithm is adopted to perform partitioning dynamic programming on the data after the initial detection to realize the detection and tracking of weak targets,thus achieving the effective detection of dense targets in complex environment.Finally,the performance of the proposed algorithm is verified by simulation and measured data,and the results show that the algorithm is efficient and robust.
作者
安政帅
AN Zhengshuai(No.20 Research Institute,China Electronics Technology Group Corporation,Xi’an 710068,China)
出处
《空天预警研究学报》
CSCD
2023年第4期252-256,共5页
JOURNAL OF AIR & SPACE EARLY WARNING RESEARCH
基金
陕西省青年科技新星项目(2022KJXX-94)。
关键词
动态规划
检测前跟踪
多目标检测
dynamic programming(DP)
track-before-detect(TBD)
multi-target detection
