局部自适应加权的逆结构稀疏表示跟踪算法

吉训生; 陈赛; 王荣飞

doi:doi:10.3969/j.issn.1003-501x.2016.12.014

光电工程, 2016, 43 (12): 85, 网络出版: 2016-12-30

局部自适应加权的逆结构稀疏表示跟踪算法

Local Adaptive Weighted Tracking Via Inverse Structure Sparse Representation

论文大纲

吉训生陈赛王荣飞

作者单位

江南大学物联网工程学院，江苏无锡 214122

稀疏表示自适应加权过完备字典候选目标 sparse representation adaptive weighting over-complete dictionary candidate target

摘要

传统的稀疏表示跟踪算法直接利用较为简单的灰度特征进行稀疏表示系数计算，易受遮挡、形变等影响。为此，提出一种局部自适应加权算法来增加受遮挡、形变等影响的候选目标与未受遮挡、形变等影响的候选目标之间的区分度。另外，一般稀疏表示算法利用数量较少的目标模板构建过完备字典。无法获得较好的稀疏系数。提出逆稀疏表示算法，利用包含丰富目标特征和背景特征的候选目标构建过完备字典来重构目标模板，相同维度的目标模板条件下可以获得更好的稀疏系数。实验表明，该算法在目标背景差异小或严重遮挡、形变情况下，都能够较好的跟踪目标。

Abstract

Traditional sparse representation tracker use simple grayscale characteristics in calculating sparse coefficient, which is easily affected by the heavy occlusions and deformation. To this end, a local adaptive weighting algorithm is put forward to increase degree of differentiation between the candidate targets affected by shade, deformation, etc and not affected by the shade, deformation, etc. In addition, the general sparse representation algorithm use a small number of target templates to build a complete dictionary, which unable to get a better sparse coefficient. Inverse structure sparse representation algorithm, using the candidate target which contains rich target and background features to build a complete dictionary to reconstruct the target template under the condition of the same dimension target template better sparse coefficient can be obtained, is proposed. Experiments show that the proposed algorithm in the small differences between target and background or serious barrier, deformation, can better track the target.

参考文献

[1] WU YI，LIM Jongwoo，YANG Ming-Hsuan. Online object tracking：A benchmark [C]// Proceedings of IEEE Conference onComputer Vision and Pattern Recognition(CVPR)，Portland，USA，June 23-28，2013：2411–2418.

[2] Doucet A，de Freitas N，Gordon N. Sequential Monte Carlo Methods In Practice [M]. New York，Springe，2001.

[3] Bradski G R. Real time face and object tracking as a component of a perceptual user interface [C]// Workshop on Applicationsof Computer Vision，Princeton，USA，Oct 19-21，1998：214-219.

[4] YANG Fan，LU Huchuan，YANG Ming-Hsuan. Robust superpixel tracking [J]. IEEE Transactions on Image Processing(S1057-7149)，2014，23(4)：1639-1651.

[5] HOANG Vandung，Le Myha，Jo Kanghyun. Hybrid cascade boosting machine using variant scale blocks based HOG featuresfor pedestrian detection [J]. Neurocomputing(S0925-2312)，2014，135(8)：357-366.

[6] 侯跃恩，李伟光，容爱琼，等. 融合背景信息的分块稀疏表示跟踪算法 [J]. 华南理工大学报(自然科学版)，2013，41(8)：21-27.

HOU Yue en，LI Weiguang，RONG Aiqiong，et al. Tracking Algorithm of Block Sparse Representation with BackgroundInformation [J]. Journal of South China University of Technology(Natural Science Edition)，2013，41(8)：21-27.

[7] JIA Xu，LU Huchuan，YANG Minghsuan. Visual tracking via adaptive structural local sparse appearance model [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR)，USA，June 16-21，2012：1822–1829.

[8] MEI Xue，LING Haibin. Robust visual tracking using l1 minimization [C]// IEEE International Conference on ComputerVision (ICCV)，Kyoto，Japan，Sept. 29-Oct. 2，2009：1436–1443.

[9] BAO Chenglong，WU Yi，LING Haibin，et al. Real time robust L1 tracker using accelerated proximal gradient approach [C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR)，USA，June 16-21，2012：1830–1837.

[10] WANG Dong，LU Huchuan，YANG Ming-Hsuan. Least soft-threshold squares tracking [C]// Proceedings of IEEE Conferenceon Computer Vision and Pattern Recognition(CVPR)，Portland，USA，June 23-28，2013：2371-2378.

[11] Freund Yoav，Schapire Robert E. A desicion-theoretic generalization of on-line learning and an application to boosting [M]Computational Learning Theory：Springer Berlin Heidelberg，1995：23-37.

[12] LIM Jongwoo，ROSS David，LIN Ruci-Sung，et al. Incremental learning for robust visual tracking [J]. International Journalof Computer Vision(S0920-5691)，2008，77(1)：125–141.

[13] ZHANG Kaihua，ZHANG Lei，YANG Ming-Hsuan. Real-time object tracking via online discriminative feature selection [J].IEEE Transactions on Image Processing(S1057-7149)，2013，22(12)：4664-4677.

[14] WebLink. SPAMS: http://www.di.ens.fr/willow/ SPAMS/downloads.html, 2015, October.

吉训生, 陈赛, 王荣飞. 局部自适应加权的逆结构稀疏表示跟踪算法[J]. 光电工程, 2016, 43(12): 85. JI Xunsheng, CHEN Sai, WANG Rongfei. Local Adaptive Weighted Tracking Via Inverse Structure Sparse Representation[J]. Opto-Electronic Engineering, 2016, 43(12): 85.

局部自适应加权的逆结构稀疏表示跟踪算法

关于本站 Cookie 的使用提示

全站搜索

局部自适应加权的逆结构稀疏表示跟踪算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索