首页 | 本学科首页   官方微博 | 高级检索  
     

一种目标轮廓跟踪的UPF方法
引用本文:袁健,张文霞,隋树林. 一种目标轮廓跟踪的UPF方法[J]. 青岛科技大学学报(自然科学版), 2006, 27(4): 355-358
作者姓名:袁健  张文霞  隋树林
作者单位:青岛科技大学,自动化与电子工程学院,山东,青岛,266042;青岛科技大学,自动化与电子工程学院,山东,青岛,266042;青岛科技大学,自动化与电子工程学院,山东,青岛,266042
摘    要:针对通用目标轮廓跟踪中CONDENSATION粒子算法的不足,提出一种轮廓跟踪的Unscented Particle Filter(UPF)方法。通过将Unscented卡尔曼滤波器与粒子滤波的结合,并实时考虑观测值,可以精确地、实时地跟踪目标区域。通过人脸跟踪结果表明,该方法对背景干扰具有很强的鲁棒性。

关 键 词:轮廓跟踪  UPF方法  鲁棒性
文章编号:1672-6987(2006)04-0355-04
修稿时间:2005-12-14

One Method of Object Contour-tracking Based on Unscented Particle Filter
YUAN Jian,ZHANG Wen-xia,SUI Shu-lin. One Method of Object Contour-tracking Based on Unscented Particle Filter[J]. Journal of Qingdao University of Science and Technology:Natutral Science Edition, 2006, 27(4): 355-358
Authors:YUAN Jian  ZHANG Wen-xia  SUI Shu-lin
Abstract:Because of the deficiency of CONDENSATION method on the object contour-tracking,one method of object contour-tracking based on unscented particle filter is proposed in the paper.The algorithm combin the Unscented Kalman filter and Particle filter with real time observed values taken into account,so real-time object contour is tracked accurately.Moreover,the method is of strong robusticity on background disturbance.
Keywords:contour tracking  Unscented Particle Filter method  robusticity
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号