基于特征在线选择的目标压缩跟踪算法

李庆武; 朱国庆; 周妍; 霍冠英

doi:10.16383/j.aas.2015.c140809

基于特征在线选择的目标压缩跟踪算法

doi: 10.16383/j.aas.2015.c140809

李庆武^1,2, ,,
朱国庆¹,
周妍^1,2,
霍冠英^1,2

1.
河海大学物联网工程学院常州 213022;
2.
常州市传感网与环境感知重点实验室常州 213022

基金项目:

国家自然科学基金(41306089),江苏省自然科学基金(BK20130240),江苏省产学前瞻性研究项目(BY2014041)资助

详细信息

作者简介:
朱国庆河海大学物联网工程学院硕士研究生.2012年获得河海大学学士学位.主要研究方向为数字图像处理,目标跟踪.E-mail:zhu_gq0902@163.com

周妍河海大学物联网工程学院讲师,博士研究生.分别于2003年和2007年在西安交通大学获得学士和硕士学位.主要研究方向为数字图像处理和模式识别.E-mail:strangeryan@163.com

霍冠英河海大学物联网工程学院副教授,2012年获得河海大学博士学位.主要研究方向为声呐图像处理.E-mail:huoguanying@hhu.edu.cn

通讯作者:
李庆武河海大学物联网工程学院教授.主要研究方向智能感知与图像处理,目标检测与跟踪.本文通信作者.E-mail:Li_qingwu@163.com

计量
- 文章访问数: 1710
- HTML全文浏览量: 73
- PDF下载量: 1630
- 被引次数: 0
出版历程
- 收稿日期: 2014-12-05
- 修回日期: 2015-05-27
- 刊出日期: 2015-11-20

Object Compressive Tracking via Online Feature Selection

LI Qing-Wu^{1,2
, ,},
ZHU Guo-Qing¹,
ZHOU Yan^1,2,
HUO Guan-Ying^1,2

1.
College of Internet of Things Engineering, Hohai University, Changzhou 213022;
2.
Changzhou Key Laboratory of Sensor Networks and Environmental Sensing, Changzhou 213022

Funds:

Supported by National Natural Science Foundation of China (41306089), Natural Science Foundation of Jiangsu Province (BK20130240), Prospective Research Project on the Integration of Industry, Education and Research of Jiangsu Province (BY2014041)

摘要

摘要: 基于压缩感知理论的压缩跟踪算法能够有效地实现对目标的跟踪, 具有良好的实时性, 但该算法对目标特征没有进行在线选择导致跟踪鲁棒性不高. 本文提出一种基于特征在线选择的目标压缩跟踪算法. 首先, 在目标附近采样得到正负样本集合, 计算样本的多尺度矩形特征, 采用压缩感知中的随机投影矩阵对高维特征投影得到低维压缩域特征, 对压缩域特征进行在线选择提取最优特征, 剔除被污染的样本特征, 使用简单高效的朴素贝叶斯分类模型进行样本判断, 实现对目标的跟踪, 同时对跟踪中目标在摄像头中的尺度变化进行建模, 给出目标尺度变化的定量描述, 实现了适应目标尺度变化的多尺度跟踪. 实验结果表明本文算法具有更好的鲁棒性与更高的跟踪精度, 对目标跟踪中的遮挡、光线突变、尺度变化和非刚性形变等因素具有较好的抗干扰能力, 同时算法复杂度低, 可以满足实时性要求.
- 特征在线选择 /
- 压缩感知 /
- 尺度变化 /
- 目标跟踪
Abstract: The compressive tracking algorithm based on compressive sensing theory can efficiently achieve real-time object tracking, but the algorithm does not select proper object features online, resulting in low tracking robustness. In order to solve this problem, an object compressive tracking algorithm with online feature selection is presented. Firstly, sets of positive and negative samples are obtained by sampling around the object, and the multi-scale rectangle features of the samples are calculated. Secondly, the compressive sensing random projection matrix is used to reduce the dimensionality of high dimensional features to obtain low-dimensional compressive domain features, and the compressive domain features are updated and selected online to extract the optimal feature to remove contaminated samples and update the classifier. Finally, a simple and efficient Bayesian classification model is utilized to achieve the object tracking. Moreover, changes of object scale in the camera are modeled and a quantitative description of changes in scale is given for multi-scale tracking which can adapt to change of the object scale. Experimental results show that the proposed algorithm can achieve a higher tracking accuracy and better robustness than several state-of-the-art algorithms and can well respond to the interferences such as block in the object tracking, light mutation, scale changes, non-rigid deformation and so on. Meanwhile, it has a low computational complexity and fully satisfies the real-time requirement.
- Online feature selection /
- compressive sensing /
- scale change /
- object tracking

HTML全文

参考文献(25)

[1]	Ross D A, Lim J, Lin R S, Yang M H. Incremental learning for robust visual tracking. International Journal of Computer Vision, 2008, 77(1-3):125-141
[2]	[2] Mei X, Ling H B. Robust visual tracking and vehicle classification via sparse representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(11):2259-2272
[3]	Yang Da-Wei, Cong Yang, Tang Yan-Dong. Object tracking method based on particle filter and sparse representation. Pattern Recognition and Artificial Intelligence, 2013, 26(7):680-687(杨大为, 丛杨, 唐延东. 基于粒子滤波与稀疏表达的目标跟踪方法. 模式识别与人工智能, 2013, 26(7):680-687)
[4]	Jiang Ming-Xin, Wang Hong-Yu, Wang Jie, Wang Biao. Visual object tracking algorithm based on ML and L2-Norm. Acta Electronica Sinica, 2013, 41(11):2307-2313(姜明新, 王洪玉, 王洁, 王彪. 基于ML和L2范数的视频目标跟踪算法. 电子学报, 2013, 41(11):2307-2313)
[5]	Qi Mei-Bin, Yang Xun, Yang Yan-Fang, Lu Lei, Jiang Jian-Guo. Real-time object tracking based on L2-norm minimization. Journal of Image and Graphics, 2014, 19(1):36-44(齐美彬, 杨勋, 杨艳芳, 陆磊, 蒋建国. 基于L2范数最小化的实时目标跟踪. 中国图象图形学报, 2014, 19(1):36-44)
[6]	Wang Zhi-Ling, Chen Zong-Hai, Xu Xiao-Xiao, Wu Liang. A fuzzy region understanding tactic for object tracking based on frog's vision characteristic. Acta Automatica Sinica, 2009, 35(8):1048-1054(王智灵, 陈宗海, 徐萧萧, 吴亮. 基于蛙眼视觉特性的运动目标模糊化区域理解跟踪方法. 自动化学报, 2009, 35(8):1048-1054)
[7]	Li Zhi-Yong, He Shuang, Liu Jun-Min, Li Ren-Fa. Motion filtering by modelling R3 cell's receptive field in frog eyes. Acta Automatica Sinica, 2015, 41(5):981-990(李智勇, 何霜, 刘俊敏, 李仁发. 基于蛙眼R3细胞感受野模型的运动滤波方法. 自动化学报, 2015, 41(5):981-990)
[8]	Li Wan-Yi, Wang Peng, Qiao Hong. A survey of visual attention based methods for object tracking. Acta Automatica Sinica, 2014, 40(40):561-576(黎万义, 王鹏, 乔红. 引入视觉注意机制的目标跟踪方法综述. 自动化学报, 2014, 40(4):561-576)
[9]	[9] Collins R T, Liu Y X, Leordeanu M. Online selection of discriminative tracking features. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(10):1631-1643
[10]	Babenko B, Yang M H, Belongie S. Robust object tracking with online multiple instance learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(8):1619-1632
[11]	Zhang K H, Zhang L, Yang M H. Real-time object tracking via online discriminative feature selection. IEEE Transactions on Image Processing, 2013, 22(12):4664-4677
[12]	Zhang K H, Zhang L, Yang M H. Real-time compressive tracking. In:Proceedings of the 12th European Conference on Computer Vision. Florence, Italy:IEEE, 2012. 864-877
[13]	Zhang K H, Zhang L, Yang M H. Fast compressive tracking. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(10):2002-2015
[14]	Achlioptas D. Database-friendly random projections:Johnson-Lindenstrauss with binary coins. Journal of Computer and System Sciences, 2003, 66(4):671-687
[15]	Baraniuk R, Davenport M, DeVore R, Wakin M. A simple proof of the restricted isometry property for random matrices. Constructive Approximation, 2008, 28(3):253-263
[16]	Ng A Y, Jordan M I. On discriminative vs. generative classifiers:a comparison of logistic regression and naive Bayes. In Advances in Neural Information Processing Systems. 2002, 14:841-848
[17]	Diaconis P, Freedman D. Asymptotics of graphical projection pursuit. The Annals of Statistics, 1984, 12(3):793-815
[18]	Kalal Z, Matas J, Mikolajczyk K. P-N learning:bootstrapping binary classifiers by structural constraints. In:Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, California, USA:IEEE, 2010. 49-56
[19]	Grabner H, Grabner M, Bischof H. Real-time tracking via on-line boosting. In:Proceedings of the 17th British Machine Vision Association. Edinburgh, UK:IEEE, 2006. 47-56
[20]	Henriques J F, Caseiro R, Martins P, Batista J. Exploiting the circulant structure of tracking-by-detection with kernels. In:Proceedings of the 12th European Conference on Computer Vision. Florence, Italy:IEEE, 2012. 702-715
[21]	Kwon J, Lee K M. Visual tracking decomposition. In:Proceedings of the 2000 IEEE Conference on Computer Vision and Pattern Recognition. San Francisco, California, USA:IEEE, 2010. 1269-1276
[22]	Hare S, Saffari A, Torr P H S. Struck:structured output tracking with kernels. In:Proceedings of the 2011 International Conference on Computer Vision. Barcelona, Spain:IEEE, 2011. 263-270
[23]	Zhong W, Lu H C, Yang M H. Robust object tracking via sparsity-based collaborative model. In:Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition. Providence, Rhode Island, USA:IEEE, 2012. 1838-1845
[24]	Collins R T. Mean-shift blob tracking through scale space. In:Proceedings of the 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Madison, Wisconsin, USA:IEEE, 2003. II-234-40
[25]	Kwon J, Lee K M. Tracking of a non-rigid object via patch-based dynamic appearance modeling and adaptive basin hopping Monte Carlo sampling. In:Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, Florida, USA:IEEE, 2009. 1208-1215