摘要
为有效解决非线性环境中的红外目标跟踪问题,提出一种自适应粒子滤波目标跟踪算法.建立了目标加权概率模型.在滤波过程中,提出双过程粒子重抽样方法,形成对抽样粒子集的自适应调节,有效地解决了粒子退化问题.用实际红外图像序列做了实验.结果表明,在非线性环境下用该方法得到的红外目标跟踪结果优于用传统粒子滤波和扩展卡尔曼滤波算法获得的结果.
For tracking infrared target in non linear condition, an adaptive particle filtering algorithm was presented. A weighted probability model of target was established. During the filtering, a resampling method based on hi-process was proposed. On basis of the method, an adaptive adjustment to sampled particle-set was formed, which resolves efficiently particles degeneration. To valuate the efficiency of the proposed algorithm,it was applied to the real infrared image sequence tracking. Experiment result shows,in non-linear condition, in comparison with traditional FP and EKF, the proposed algorithm have great advantages in the field.
出处
《光子学报》
EI
CAS
CSCD
北大核心
2009年第6期1507-1511,共5页
Acta Photonica Sinica
基金
国家教育部博士点基金(20040699015)资助
关键词
目标跟踪
粒子滤波
双过程
加权颜色概率分布
Target tracking
Adaptive particle filter
Bi process
Weighted color probability contribution
