摘要
针对粒子滤波重采样中粒子贫化问题,采用了权值选择的优化方法,对每个粒子的权值进行排序,选取其中权值较大的粒子参与跟踪估计,使权值较小的粒子有机会参与下一状态的估计,保证参与状态估计的大部分粒子具有多样性,有效克服粒子贫化现象。为了进一步提高跟踪性能,根据红外目标成像特点,融合目标梯度特征和灰度特征建立观测模型,并根据置信度实时调整每个特征对跟踪结果的影响,且自适应更新模板。经仿真验证,红外目标在复杂背景或遇到遮挡情况下,该算法能够精确鲁棒地跟踪目标。
To deal with the problem of sample degeneration and sample impoverishment in traditional particle filter, a weight selected method is presented. The proposal distribution for each particle involves sampling from the state-space model a number of times. In order to improve the accuracy of infrared object tracking,the grey-scale feature and gradi- ent feature are combined to establish observation model! based on infrared image characteristics, and the influence of each characteristic on tracking is adjusted by depending on confidence. Simulation results show that the improved algo- rithm is robust, and can track infrared object stably under complex background.
出处
《激光与红外》
CAS
CSCD
北大核心
2014年第1期35-40,共6页
Laser & Infrared
基金
国家自然科学基金(No.61171155)
陕西省自然科学基金(No.2012JM8010)
西北工业大学研究生创业种子基金(No.Z2013070)资助
关键词
多特征
粒子滤波
红外目标
权值选择
抗遮挡
multiple features
particle fiher
infrared object
weight selected
anti-occlusion
