English

应用科学学报 >

2015 , Vol. 33 >Issue 5: 502 - 517

DOI: https://doi.org/10.3969/j.issn.0255-8297.2015.05.005

信号与信息处理

基于支持向量机的多特征选择目标跟踪

展开
  • 1. 南京信息工程大学电子与信息工程学院, 南京 210044;
    2. 南京信息工程大学江苏省大气环境与装备技术协同创新中心, 南京 210044;
    3. 中国人民解放军 94804部队, 上海 200434;
    4. 苏州电器科学研究院股份有限公司, 江苏苏州 215000
胡昭华,博士,副教授,研究方向:视频目标跟踪、模式识别,E-mail:zhaohua_hu@163.com

收稿日期: 2015-04-21

  修回日期: 2015-05-26

  网络出版日期: 2015-09-30

基金资助

国家自然科学基金(No.61203273);江苏省自然科学基金(No.BK20141004);江苏省普通高校自然科学研究基金(No.11KJB510009, No.14KJB510019);江苏省高校优势学科Ⅱ期建设工程项目基金资助

收起

Multi-feature Selection Tracking Based on Support Vector Machine

Expand
  • 1. School of Electronic & Information Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China;
    2. Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing 210044, China;
    3. 94804 PLA Troops, Shanghai 200434, China;
    4. Suzhou Electrical Apparatus Science Research Institute Co., Ltd, Suzhou 215000, Jiangsu Province, China

Received date: 2015-04-21

  Revised date: 2015-05-26

  Online published: 2015-09-30

Fold

摘要

目标判别跟踪方法一般以单一跟踪结果(正样本)和部分负样本训练分类器,并只用单一的特征描述目标,因此在目标发生遮挡、光照变化、形变等情况下很容易导致跟踪的失败. 为了克服单一正样本和单一特征训练的分类器导致跟踪不稳定的问题,提出一种基于支持向量机的多特征描述、多个正负样本进行分类器训练的目标跟踪方法,用训练分类器对所有候选粒子的不同特征进行判别和选择,并将具有最大置信概率的粒子确定为跟踪结果. 在跟踪过程中,利用子空间学习的方法实现正样本的更新,并以更新后的正负样本训练分类器. 实验结果表明,该方法在目标存在遮挡、光照变化、形变等情况下均可以取得较好的跟踪效果.

关键词： 分类器; 目标跟踪; 支持向量机; 子空间学习; 多特征选择

本文引用格式

胡昭华, 徐玉伟, 赵孝磊, 何军, 周游 . 基于支持向量机的多特征选择目标跟踪[J]. 应用科学学报, 2015 , 33(5) : 502 -517 . DOI: 10.3969/j.issn.0255-8297.2015.05.005

Abstract

Discriminative tracking is generally based on a single feature and uses the current track result (a positive sample) and some negative samples to train the classifier. It may lead to tracking due to occlusion, illumination changes and deformation. To overcome the problem caused by single feature description and single positive sample training, we propose a multi-feature selection algorithm based on the support vector machine (SVM). The classifier is trained with multiple positive and negative samples based on multi-feature descriptions. In the tracking step, a candidate sample with maximum confidence probability is used as the tracking result. Further, we take advantage of a subspace learning method to update positive samples used in the classifier training. Experimental results show that the proposed algorithm has desirable performance in a variety of challenging situations.

Key words： classifier; object tracking; support vector machine (SVM); subspace learning; multi-feature selection

参考文献

[1] Yilmaz A, Javed O, Shah M. Object tracking: a survey [J]. ACM Computing Surveys, 2006, 38(4): 1-45.

[2] Zhang S P, Yao H X, Sun X. Sparse coding based visual tracking: review and experimental comparison [J]. Pattern Recognition, 2013, 46 (7): 1772-1788.

[3] Wang Z H, Yoon S, Xie S J, Lu Y, Rark D S. Visual tracking with semi-supervised online weighted multiple instance learning [J]. The Visual Computer, 2015: 1-14.

[4] Yi K M, Jeong H, Lee B, Choi J Y. Visual tracking in complex scenes through pixel-wise tri-modeling [J]. Machine Vision and Applications, 2015: 1-13.

[5] Yang X, Wang M, Tao D C. Robust visual tracking via multi-graph ranking [J]. Neurocomputing, 2015.

[6] Zhong B N, Shen Y J, Chen Y. Online learning 3D context for robust visual tracking [J]. Neurocomputing, 2015, 151: 710-718.

[7] Ma C, Liu C C. Two dimensional hashing for visual tracking [J]. Computer Vision and Image Understanding, 2015.

[8] Li H X, Li Y, Porikli F. Deep track: learning discriminative feature representations online for robust visual tracking [J]. arXiv preprint arXiv: 1503. 00072, 2015.

[9] Hwang J P, Baek J, Choi B, Kim E. A novel part-based approach to mean-shift algorithm for visual tracking [J]. International Journal of Control, Automation and Systems, 2015: 1-11.

[10] Wang L J, Lu H C, Wang D. Visual tracking via structure constrained grouping [J]. Signal Processing Letters, 2015, 7(22): 794-798.

[11] Zeng F X, Liu X, Huang Z T, Ji Y F. Kernel based multiple cue adaptive appearance model for robust real-time visual tracking [J]. IEEE Signal Processing Letters, 2013, 20(11): 1094-1097.

[12] Yu Q, Dinh T B, Medioni G. Online tracking and reacquistion using co-trained generative and discriminative trackers [C]//Proceedings of 10th European Conference on Computer Vision, Marseille: Springer Berlin Heidelberg, 2008, 5303: 678-691.

[13] Bai T X, Li Y F. Robust visual tracking with structured sparse representation appearance model[J]. Pattern Recognition, 2012, 45(6): 2390-2404.

[14] Ross D A, Lim J, Lin R S, Yang M H. Incremental learning for robust visual tracking [J]. International Journal of Computer Vision, 2008, 77 (1/3): 125-141.

[15] Mei X, Ling H B. Robust visual tracking using L1 minimization [C]//Proceedings of 2009 IEEE 12th International Conference on Computer Vision, Kyoto: IEEE, 2009: 1436-1443.

[16] Wang D, Lu H C, Yang M H. Online object tracking with sparse prototypes [J]. IEEE Transctions on Image Processing, 2013, 22(1): 314-325.

[17] Zhang T Z, Ghanem1 B, Liu S, Ahuja N. Low-rank sparse learning for robust visual tracking[C]// Proceedings of 12th European Conference on Computer Vision, Florence:, Springer Berlin Heidelberg, 2012, 7577: 470-484.

[18] Grabner H, Leistner C, Bischof H. Semi-supervised on-line boosting for robust tracking[C]//Proceedings of 10th European Conference on Computer Vision, Marseille: Springer Berlin Heidelberg, 2008, 5302: 234-247.

[19] Babenko B, Yang M H, Belongie S. Visual tracking with online multiple instance learning[C]//Proceedings of 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami: IEEE, 2009: 983-990.

[20] Zhang K H, Zhang L, Yang M H. Real-time compressive tracking [C]//Proceedings of 12th European Conference on Computer Vision, Florence:, Springer Berlin Heidelberg, 2012: 864-877.

[21] Zhang K H, Zhang L, Yang M H. Real-time object tracking via online discriminative feature selection [J]. IEEE Transactions on Image Processing, 2013, 22(12): 4664-4677.

[22] Avidan S. Support vector tracking [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26: 1064-1072.

[23] Collins R, Liu Y X. Leordeanu M. Online selection of discriminative tracking features [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27: 1631-1643.

[24] Grabner H, Grabner M, Bischof H. Real-time tracking via online boosting [C]//Proceedings of 2006 Proceedings of the British Machine Vision Conference, Edinburgh, 2006.

[25] Gammermann A. Support vector machine learning algorithm and transduction [J]. Computational Statistics, 2000, 15(1): 31-39.

[26] Jia X, Lu H C, Yang M H. Visual tracking via adaptive structural local sparse appearance model[C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1822-1829.

[27] Ojala T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971-987.

[28] Khan Z, Batch T, Dellaert F. A Rao–blackwellized particle filter for eigen-tracking[C]//Proceedings of 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Washington, 2004, 2: 980-986.

[29] Zhou S K, Chellappa R, Moghaddam B. Visual tracking and recognition using appearanceadaptive models in particle filters [J]. IEEE Transactions on Image Processing, 2004, 13(11): 1491-1506.

[30] Yuan X T, Liu X B, Yan S C. Visual classification with multitask joint sparse representation[J]. IEEE Transactions on Image Processing, 2012, 21(10): 4349-4360.

[31] John C P. Sequential minimal optimization: a fast algorithm for training support vector machines[J]. [S.l.]: MIT Press, 1998: 1-21.

[32] Olson C F. Maximum-likelihood template matching [C]//Proceedings of 2000 IEEE Conference on Computer Vision and Pattern Recognition, Hilton Head Island, 2000, 2: 52-57.

[33] Matthews I, Ishikawa T, Baker S. The template update problem [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26 (6): 810-815.

[34] Bao C L, Wu Y, Ling H B. Real time robust `1 tracker using accelerated proximal gradient approach [C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, IEEE, 2012: 1830-1837.

[35] Kalal Z, Mikolajczyk K, Matas J. Tracking-learning-detection [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7): 1409-1422.

[36] Kwon J, Lee K M. Visual tracking decomposition [C]//Proceedings of 2010 IEEE Conference on Computer Vision and Pattern Recognition, San Francisco, 2010: 1269-1276.

[37] Adam A, Rivlin E, Shimshoni T. Robust fragments-based tracking using the integral histogram[C]//Proceedings of 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2006, 1: 798-805.

[38] Zhang T Z, Ghanem B, Liu S, Et A L .Robust visual tracking via structured multi-task sparse learning [J]. International Journal of Computer Vision, 2013, 101(2): 367-383.

[39] Liu B Y, Huang J Z, Yang L, Kulikowsk C. Robust tracking using local sarse appearance model and k-selection [C]//Proceedings of 2011 IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2011: 1313-1320.

[40] Zhong W, Lu H C, Yang M H. Robust object tracking via sparsity-based collaborative model[C]//Proceedings of 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, 2012: 1838-1845.

[41] Wu Y, Lim J, Yang M H. Online object tracking: a benchmark [C]//2013 IEEE Conference on Computer Vision and Pattern Recognition, 2013: 2411-2418.
Options
文章导航

/

〈 〉