[1]林渊灿,陈锻生,胡小平.嵌入尺度可变均值漂移算法的粒子滤波方法[J].华侨大学学报(自然科学版),2010,31(4):408-412.[doi:10.11830/ISSN.1000-5013.2010.04.0408]
　LIN Yuan-can,CHEN Duan-sheng,HU Xiao-ping.A Particle Filter Method Embedded with a Variable Scale Mean-Shift Algorithm[J].Journal of Huaqiao University(Natural Science),2010,31(4):408-412.[doi:10.11830/ISSN.1000-5013.2010.04.0408]
点击复制

嵌入尺度可变均值漂移算法的粒子滤波方法()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第31卷
期数:
2010年第4期
页码:
408-412
栏目:
出版日期:
2010-07-20

文章信息/Info

Title:
A Particle Filter Method Embedded with a Variable Scale Mean-Shift Algorithm
文章编号:
1000-5013(2010)04-0408-05
作者:
林渊灿; 陈锻生; 胡小平
华侨大学计算机科学与技术学院
Author(s):
LIN Yuan-can;  CHEN Duan-sheng;  HU Xiao-ping
College of Computer Science and Technology, Huaqiao University, Quanzhou 362021, China
关键词:
粒子滤波;  均值漂移算法;  尺度可变;  对数极坐标;  目标跟踪
Keywords:
particle filter;  mean-shift;  variable scale;  logarithmic polar coordinate;  object tracking
分类号:
TP391.41
DOI:
10.11830/ISSN.1000-5013.2010.04.0408
文献标志码:
A
摘要:
将尺度可变均值漂移算法嵌入到粒子的扩散过程中,引导粒子扩散到后验概率密度函数的高密度区,提出一种嵌入尺度可变均值漂移算法的粒子滤波跟踪方法.利用对数极坐标图像的尺度不变性,在粒子扩散过程中同时进行位置、尺度空间漂移.实验表明,该方法不仅能顺利跟踪非连续尺度变化目标,而且需要更少的粒子数.
Abstract:
Embedding variable scale mean-shift algorithm into particle diffusion process,particles are diffusion into the high density area of the post probability density function,this paper proposed a particle filter tracking method embedded with a variable scale mean-shift algorithm.The scale invariable characters of image in logarithmic polar coordinates was used in the particle diffusion process,in which position space and scale space are shifted at the same time.The experiment shows the method can not only track object smoothly with discontinuous scale variation,but also need less particles.

参考文献/References:

[1] COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003(5):564-577.doi:10.1109/TPAMI.2003.1195991.
[2] ISARD M, BLAKE A. Condensation:Conditional density propagation for visual tracking [J]. International Journal of Computer Vision, 1998(1):5-28.doi:10.1023/A:1008078328650.
[3] PENG Ning-song, YANG Jie, LIU Zhi. Performance analysis for tracking of variable scale objects using mean-shift algorithm [J]. Optical Engineering, 2005(7):070505.doi:10.1117/1.1985487.
[4] MAGGIO E, CAVALLARO A. Hybrid particle filter and mean shift tracker with adaptive transition model [A]. Philadelphia:IEEE, 2005.19-23.
[5] HAN R, JING Z, LI Y. Kernel based visual tracking with variant spatial resolution model [J]. Electronics Letters, 2008(8):517-518.doi:10.1049/el:20080051.
[6] SCHWARTZ E L. Anatomical and physiological correlates of visual computation from striate to infero-temporal cortex [J]. IEEE Transactions on Systems Man and Cybernetics, 1984(2):257-271.
[7] KAILATH T. The divergence and bhattacharyya distance measures in signal delection [J]. IEEE Transactions on Communications Technology, 1967(1):52-60.doi:10.1109/TCOM.1967.1089532.
[8] NUMMIARO K, KOLLER-MEIERB E, VAN GOOL L. An adaptive color-based particle filter [J]. Image and Vision Computing, 2003(1):99-110.doi:10.1016/S0262-8856(02)00129-4.
[9] 张波. 基于粒子滤波的图像跟踪算法研究 [D]. 上海:上海交通大学, 2007.

备注/Memo

备注/Memo:
福建省科技计划重点项目(2008I0021); 福建省自然科学基金计划资助项目(2009J01289)
更新日期/Last Update: 2014-03-23