摘要
目标数未知或随时间变化是红外弱小目标跟踪技术的一个难题。为解决这个问题,提出了基于概率假设密度滤波的红外弱小目标跟踪算法。从数据关联的角度出发,将目标集看作随机集,利用概率假设密度滤波的数据关联算法实现目标数未知的红外弱小目标的跟踪。实验结果表明,在杂波环境下,概率假设密度滤波可以稳健地跟踪红外弱小目标的目标状态和目标数目。
The number of objects to track is unknown or varies over time is a difficult problem for infrared dim and small target tracking technique. The algorithm of infrared dim and small target tracking based on probability hy- pothesis density （PHD） filter is presented. From the data associated with the perspective of the target set as a ran- dom set, using the probability hypothesis density filter to achieve the target data association algorithm for infrared dim and small target with an unknown object. Experiments show the PHD filter to be able to estimate both the number of infrared dim and small tracked objects and the state of the objects, robustly in cluttered environment.
出处
《科学技术与工程》
2010年第12期2894-2896,2901,共4页
Science Technology and Engineering
基金
渭南师范学院研究生专项科研项目(09YKZ015)资助
关键词
多目标跟踪
概率假设密度(PHD)滤波
红外图像
目标检测
multi-target tracking probability hypothesis density （PHD） filter infrared image target detection
