|
|
论文:2012,Vol:30,Issue(3):457-460 |
|
|
引用本文： |
|
|
张君昌, 牛步杨. 基于有效特征筛选的Meanshift运动目标跟踪算法[J]. 西北工业大学 |
|
|
Zhang Junchang, Niu Buyang. A New and Effective Mean Shift Object Tracking Algorithm Based on Effective Feature Extraction[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 基于有效特征筛选的Meanshift运动目标跟踪算法 |
|
| 张君昌, 牛步杨 |
|
| 西北工业大学 电子信息学院,陕西 西安 710072 |
| 摘要: |
| 针对现有Mean shift跟踪算法在目标被遮挡、跟踪场景变化时,跟踪误差变大甚至丢失目标的问题,提出了一种基于有效特征筛选的Mean shift运动目标跟踪算法。首先通过对目标特征的优化筛选,改善了现有Mean shift算法因目标特征多而造成计算时间较长,在目标发生较大变化时跟踪精度降低的情况。更能有效地表征目标特征,减少跟踪误差,增强特征集对目标的描述能力。同时给出目标模板更新的方法,在目标发生明显变化时,能自适应地更新特征集,进一步提高跟踪精度。仿真结果表明:文中方法具有更好的跟踪精度,计算时间较小,对遮挡、场景变化有更好的鲁棒性。 |
| 关键词:
Mean shift算法
目标跟踪
特征选取
模板更新
|
|
| A New and Effective Mean Shift Object Tracking Algorithm Based on Effective Feature Extraction |
|
| Zhang Junchang, Niu Buyang |
|
| Department of Electronics Engineering,Northwestern Polytechnical University,Xi'an 710072,China |
| Abstract: |
| Aim. The introduction of the full paper reviews a number of papers in the open literature and then pro-poses what we believe to be a new and effective algorithm,which is explained in sections 1 and 2. Section 1 briefsthe traditional Mean shift algorithm. The core of section 2 consists of: (1) our new and effective feature extractionalgorithm improves the present Mean shift tracking algorithms in that the calculation time is shorter and that thetracking accuracy is higher; the target’ s features can be displayed more effectively,tracking errors can be re-duced,and the description of the features set can be enhanced; eqs. (10) and (11) are worth particular attention;(2) the target updating method we propose can adaptively update feature set when the target changes greatly andsuddenly,and enhance tracking accuracy; eqs. (12),(13),and (14) are worth particular attention. Simulationresults,presented in Figs. 1 through 4,and their analysis show preliminarily that our new effective feature extrac-tion algorithm has indeed higher localization precision and requires indeed less computational time. |
| Key words:
algorithms
computational efficiency
efficiency
error analysis
feature extraction
targets
tracking(position) ;Mean shift algorithm
template updating
|
|
| 收稿日期: 2011-06-06
修回日期:
|
| DOI: |
| 通讯作者:
Email： |
| 作者简介: 张君昌(1969-),西北工业大学副教授,主要从事信号处理和天线通信研究。
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Comaniciu D，Ramesh V，Meer P.Kernel-Based Object Tracking.IEEE Trans on Pattern Analysis and Machine Intelligence，2005，25(5): 564-577
[2] Kim T，Lee S，Paik J.Combined Shape and Feature-Based Video Analysis and Its Application to Non-Rigid Object Tracking.IET Image Process，2011，5(1): 87-100
[3] Wang Junqiu，Yagi Yasushi.Integrating Color and Shape-Texture Features for Adaptive Real-Time Object Tracking.IEEETrans on Image Processing，2008，17(2): 235-240
[4] Amir Babaeian，Saeed Rastegar，Mojtaba Bandarabadi，Maziar Rezaei.Mean Shift-Based Object Tracking With Multiple Fea-tures.IEEE Trans on Pattern Analysis and Machine Intelligence，March 2009，27(10): 78-80
[5] Chen Qiang，Sun Quansen，Heng Phengann.Two Stage Object Tracking Method Based on Kernel and Active Contour.IEEETrans on Circuits and System for Video Technology，2010，20(4): 605-609 |
|
|
|
|
|
|
|
