[1]王业祥,朱文球,孙文静.局部加权距离度量的双向稀疏表示目标跟踪[J].计算机技术与发展,2018,28(04):60-64.[doi:10.3969/ j. issn.1673-629X.2018.04.013]
　WANG Ye-xiang,ZHU Wen-qiu,SUN Wen-jing.Target Tracking of Bidirectional Sparse Representation of Local Weighted Distance Metric[J].,2018,28(04):60-64.[doi:10.3969/ j. issn.1673-629X.2018.04.013]
点击复制

局部加权距离度量的双向稀疏表示目标跟踪()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年04期
页码:
60-64
栏目:
智能、算法、系统工程
出版日期:
2018-04-10

文章信息/Info

Title:
Target Tracking of Bidirectional Sparse Representation of Local Weighted Distance Metric
文章编号:
1673-629X(2018)04-0060-05
作者:
王业祥1   2 ; 朱文球1   2 ; 孙文静1   2
1. 湖南工业大学 计算机学院,湖南 株洲 412007;
2. 智能信息感知及处理技术湖南省重点实验室,湖南 株洲 412007
Author(s):
WANG Ye-xiang 1; 2 ; ZHU Wen-qiu 1; 2 ; SUN Wen-jing 1; 2
1. School of Computer,Hunan University of Technology,Zhuzhou 412007,China；
2. Key Laboratory of Intelligent Information Perception and Processing Technology (Hunan Province),Zhuzhou 412007,China
关键词:
视觉跟踪; 双向稀疏; L 1 范数; 加速逼近梯度; 局部权重距离度量
Keywords:
visual tracking; bidirectional sparse; L1-norm; accelerating approximation gradient; locally weighted distance metric
分类号:
TP31
DOI:
10.3969/ j. issn.1673-629X.2018.04.013
文献标志码:
A
摘要:
首先提出基于双向稀疏表示的目标跟踪模型框架。 该模型用 L 1 范数来约束正反向重构误差,通过利用加速逼近梯度(APG)算法求得正反稀疏系数矩阵,根据目标正负模板集和候选模板集之间的距离度量得到权重矩阵。 通过权重矩阵与正反稀疏系数矩阵,得到候选样本集中正负差异度最大的候选样本,把最优候选样本作为跟踪最优目标；然后在目标模板集和候选样本集之间的距离度量上,由于传统欧氏距离权重在目标发生遮挡、光照等情况下具有不准确性,基于此提出改进的局部权重距离度量方法。 该算法在复杂环境视频序列下,相比传统目标跟踪算法具有较高的鲁棒性。
Abstract:
We first propose a framework about target tracking model based on bidirectional sparse representation,which uses the L 1 minimization to constrain the forward and reverse reconstruction errors. The positive and negative sparse coefficient matrices are obtained by the algorithm of accelerated approximation gradient (APG). In this paper,we obtain the weight matrix according to the distance between the positive and negative template set and the candidate template set. By using the weight matrix and the positive and negative sparse coefficient matrix,the candidate samples with the largest positive and negative difference between the candidate samples are obtained,of which the optimal candidate samples are selected as the tracking optimal target. And then on the distance between the target template set and the candidate sample set,because the traditional Euclidean distance is inaccurate in the case of occlusion and illumination of the target,we propose an improved local weight distance measurement method. Compared with the traditional target tracking algorithm,the proposed algorithm has high robustness in the complex environment video sequence.

相似文献/References:

[1]田应仲,刘伊芳,李 龙.联合判别式核相关滤波视觉跟随算法[J].计算机技术与发展,2021,31(03):78.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 013]
　TIAN Ying-zhong,LIU Yi-fang,LI Long.Visual Following Algorithm of Joint Discriminant Kernel Correlation Filter[J].,2021,31(04):78.[doi:10. 3969 / j. issn. 1673-629X. 2021. 03. 013]
[2]梁文辉,宋 涛,叶永达.间隔贝叶斯估计和空时目标运动边界视觉跟踪[J].计算机技术与发展,2021,31(09):75.[doi:10. 3969 / j. issn. 1673-629X. 2021. 09. 013]
　LIANG Wen-hui,SONG Tao,YE Yong-da.Visual Tracking Based on Interval Bayesian Estimation and Spatial-temporal Object Motion Boundaries[J].,2021,31(04):75.[doi:10. 3969 / j. issn. 1673-629X. 2021. 09. 013]

更新日期/Last Update: 2018-06-06