基于多特征融合的粒子滤波算法的研究与实现

被引:4
作者
刘进
陈玮
机构
[1] 上海理工大学
关键词
目标跟踪; 多特征融合; 粒子滤波; 卡尔曼滤波;
D O I
10.16526/j.cnki.11-4762/tp.2013.05.061
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
在目标被遮挡、目标颜色与背景颜色相近、目标的形状发生变化以及目标的快速移动等复杂的外部环境下,针对目标被跟丢的情况,提出了一种基于多特征融合的粒子滤波算法;基于目标的颜色和边缘特征信息的融合设计了一种自适应特征融合观测模型,同时采用卡尔曼滤波的方法快速预测下一时刻目标的运动轨迹,最后结合粒子滤波可以有效的提高在目标在各种复杂情况下的跟踪能力;实验结果表明,该算法适应性更强,在复杂环境下有更好的跟踪效果。
引用
收藏
页码:1307 / 1309
页数:3
相关论文
共 11 条
[1]  
Finding Main Road Seeds Basedon Symmetrical Edge Orientation Histogram. Tang L,Xie W X,Jianjun Huang. Electronics Letters . 2004
[2]  
Visual tracking andrecognition using appearance-adaptive models in particle filters. Zhou S.K,Chellappa R,Moghaddam B. IEEE Transactions on Image Processing . 2006
[3]  
Integrating Shape and Color Features forAdaptive Real-time Object Tracking. Wang T Q,Yasushi Y. IEEE InternationalConference on Roboics and Biomimetics . 2010
[4]  
Integral histogram:A fast way to extract histograms inCartesian spaces. Porikli F. CVPR05 . 2005
[5]  
Particle filters for continuous likelihood evaluation and maximisation. Journal of Econometrics . 2011
[6]  
A nonlinear filter coupled with hospitability and synthetic inclination maps for in-surveillance and out-of-surveillance tracking. Kassas Z. M,Ozguner U. IEEE Transactions on System, Man, and Cybernetics-Part C:Application and Reviews . 2010
[7]   基于空间直方图的多目标粒子滤波跟踪 [J].
王勇 ;
王典洪 .
光电工程, 2010, 37 (01) :65-69+75
[8]  
Monte Carlo Visual Tracking Using ColorHistograms and a Spatially Weighted Oriented Hausdorff Measure. Tao X,Christion D. Proceedings of the Conference on Analysis of Images andPatterns . 2011
[9]  
The divergence and Bhattacharyaa distance measures insignal selectio. Kailath T. IEEE Transaction on Communication Technol-ogy . 2010
[10]  
Edge orientation-based multiview objectrecognition. Zhu W,Levinson S E. ICPR00 . 2008