基于局部分块学习的在线视觉跟踪

doi:10.3969/j.issn.0372-2112.2015.01.012

PDF(3584 KB)

电子学报 ›› 2015, Vol. 43 ›› Issue (1) : 74-78. DOI: 10.3969/j.issn.0372-2112.2015.01.012

学术论文

基于局部分块学习的在线视觉跟踪

余旺盛¹, 田孝华¹, 侯志强¹, 查宇飞²

作者信息 +

Online Visual Tracking Based on Local Patch Learning

YU Wang-sheng¹, TIAN Xiao-hua¹, HOU Zhi-qiang¹, ZHA Yu-fei²

Author information +

文章历史 +

摘要

视觉跟踪中,如何构建一种能够适应目标表观特征变化的目标模型是增强算法跟踪精度和稳定性的关键之一.本文提出利用跟踪区域内像素的初始分类标记来构建目标的局部分块模型,并在贝叶斯理论框架下提出了基于局部分块学习的在线视觉跟踪算法.首先,利用标定的初始跟踪区域构建目标的局部分块模型;然后,在当前跟踪区域中通过局部分块学习和贝叶斯估计确定当前帧的跟踪结果;最后,利用特征聚类对局部分块模型进行更新.实验结果表明：所提算法对目标表观变化的适应性明显增强,跟踪精度和稳定性较近年来的同类算法均有一定提高.

Abstract

In visual tracking,how to construct an object model to cope with the appearance change is one of the key problems to improve tracking precision and stability.To resolve this problem,this paper proposes to construct a local patch model using the initial labels of the pixels in tracking area,and proposes an online visual tracking algorithm based on local patch learning under the framework of Bayesian theory.The detailed operation is as follows.Firstly,it constructs the local patch model according to the initialized tracking area.Then,it utilizes the object model to learn the local patches in current tracking area and estimates the current state via Bayes estimation.Finally,it updates the local patch model by feature clustering.The experiment results indicate that the proposed algorithm obtains a distinct improvement in coping with appearance change,and exceeds the recent local patch-based trackers in both tracking precision and stability.

导出引用

余旺盛, 田孝华, 侯志强, 查宇飞. 基于局部分块学习的在线视觉跟踪[J]. 电子学报, 2015, 43(1): 74-78. https://doi.org/10.3969/j.issn.0372-2112.2015.01.012

YU Wang-sheng, TIAN Xiao-hua, HOU Zhi-qiang, ZHA Yu-fei. Online Visual Tracking Based on Local Patch Learning[J]. Acta Electronica Sinica, 2015, 43(1): 74-78. https://doi.org/10.3969/j.issn.0372-2112.2015.01.012

中图分类号： TP391.41

参考文献

[1] Zeng Huanzhao,Peter G Anderson.Extended Neugebauer model for printer color formation[A].Proceedings of the Color Imaging:Device-Independent Color,Color Hardcopy,and Graphic Arts IV[C].USA:SPIE,1999.27-36.
[2] C De M Bezerra,C J Hawkyard.Color matching for ink-jet prints on paper[J].Color Research and Application,1998,23(1):18-26.
[3] David Littlewood,Ganesh,Subbarayan.Maintaining an accurate printer characterization[A].Proceedings of the Color Imaging Conference:Color Science,System,and Applications[C].USA:IS&T,2004.203-210.
[4] 刘瑞华,曾平,王义峰.一种自适应分区回归的打印机色彩校正方法[J].电子学报,2007,35(11):2201-2204. Liu Rui-hua,Zeng Ping,Wang Yi-feng.A color calibration method of printer using adaptive regression[J].Acta Electronica Sinica,2007,35(11):2201-2204.(in Chinese)
[5] 苏晓红,郭茂祖.基于偏最小二乘回归分析的混色数据学习算法研究[J].电子学报,2001,29(3):429-431. Su Xiaohong,Guo Maozu,et al.Research of mix-color data learning algorithm based on partial leas-t square regression[J].Acta Electronica Sinica,2001,29(3):429-431.(in Chinese)
[6] 王义峰,曾平.基于色域划分的多通道打印机色彩校正[J].电子学报,2010,38(3):507-511. Wang Yi-feng,Zeng Ping.Multi-channel printer characterization based on gamut partition[J].Acta Electronica Sinica,2010,38(3):507-511.(in Chinese)
[7] 朱明,刘真,陈广学.基于分区纽阶堡方程的6色印刷分色模型研究[J].光学学报,2011,31(7):1-10. Zhu Ming,Liu Zhen,Chen Guangxue.Research on six-color separation model based on subarea neugebauer equations[J].Acta Optica Sinica,2011,31(7):1-10.(in Chinese)
[8] 史瑞芝,曹朝辉.基于7色高保真彩色印刷的颜色分色模型[J].测绘科学,2007,32(5):58-60. Shi Ruizhi,Cao Chaohui.A color separation model based on 7-color Hi-Fi press[J].Science of Surveying and Mapping,2007,32(5):58-60.(in Chinese)
[9] 张桢杰,等.基于子空间划分的多项式回归多色分色算法[J].包装工程,2013,34(7):65-67. Zhang Zhenjie,et al.Polynomial regression multi-color separation method based on subspace partition[J].Packaging Engineering,2013,34(7):65-67.(in Chinese)