A Mean Tensor Component Analysis and Its Application in Gait Recognition

doi:10.7652/xjtuxb201512007

您当前的位置：

首页 >

文章列表页 >

A Mean Tensor Component Analysis and Its Application in Gait Recognition

更新时间：2025-07-09

- A Mean Tensor Component Analysis and Its Application in Gait Recognition
- Vol. 49, Issue 12, Pages: 40-46(2015)
- 作者机构：
  
  1. 南昌航空大学江西省图像处理与模式识别重点实验室,南昌,330063
  2. 山东大学信息科学与工程学院,济南,250100
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb201512007
  CLC： TP391.41
- Online First：10 December 2015，
  
  Published：2015
- 稿件说明：
移动端阅览
A Mean Tensor Component Analysis and Its Application in Gait Recognition[J]. 2015, 49(12): 40-46.
DOI：

A Mean Tensor Component Analysis and Its Application in Gait Recognition[J]. 2015, 49(12): 40-46. DOI： 10.7652/xjtuxb201512007.

摘要

针对步态识别中非负样本数据存在冗余

且未经中心化的多线性主成分分析保持聚类结构的特征向量不能对应最大特征值

导致识别效果下降的问题

提出一种保持原始张量数据均方长度的均值张量成分分析算法。该算法首先对原始样本任一模式下内积矩阵进行谱分解

计算该模式下相应的特征值和特征向量; 其次

利用获得的特征值和特征向量计算均值向量

并对均值向量值进行降序排列

使较大均值向量值对应的特征向量构成该模式下的低维子空间; 最后

将原始样本投影到该低维子空间

形成特征张量。与多线性主成分分析算法相比

该算法不需要对数据去中心化处理

而是保持非负数据均值向量最大均方欧几里德距离和方向。通过在USF步态数据库和TUM GAID步态数据库进行仿真实验

结果表明

经过均值张量成分分析预处理

在2个步态库上的平均识别率分别高达57%和75%

较其他传统方法的识别率有明显提高。

Abstract

A novel algorithm named mean tensor component analysis(MTCA)is proposed to solve the low recognition accuracy problem that is caused by redundancy in nonnegative sample data and non-corresponding relationship that the eigenvector which preserves clustering structure in uncentered multilinear principal component analysis(MPCA)does not correspond to the maximum eigenvalues. The algorithm reserves the squared length of original tensor data. Spectral decomposition is performed to the inner-product matrix of original samples in any mode to obtain eigenvalues and corresponding eigenvectors. Then

mean value and mean vector are calculated from the eigenvalues and eigenvectors

and the values of the mean vector are sorted in a descending order so that a subspace is formed from the eigenvectors corresponding to first several largest values of the mean vector. Then a feature tensor is acquired by mapping the original sample to the subspace. A comparison to MPCA shows that the proposed algorithm preserves the squared length and the direction of mean vector of non-negative data without needs of decentralized processing to the data. Experiments on USF and TUM GAID gait databases show that the MTCA algorithm achieves average recognition rate 57% and 75%

respectively

and that the rates are obviously higher than the recognition rates of some conventional methods.

关键词

Keywords

references

陈实, 高有行. 一种轮廓变化图像小波矩的步态识别 [J]. 西安交通大学学报, 2009, 43(1): 90-94.

CHEN Shi, GAO Youxing. Gait recognition with wavelet moments of silhouette change images [J]. Journal of Xi'an Jiaotong University, 2009, 43(1): 90-94.

贲晛烨, 徐森, 王科俊. 行人步态的特征表达及识别综述 [J]. 模式识别与人工智能, 2012, 25(1): 71-81.

BEN Xianye, XU Sen, WANG Kejun. Review on pedestrian gait feature expression and recognition [J]. Pattern Recognition Artificial Intelligence, 2012, 25(1): 71-81.

LU H, PLATANIOTIS K N, VENETSANOPOULOS A N. MPCA: multi-linear principal component analysis of tensor objects [J]. IEEE Transactions on Neural Networks, 2008, 19(1): 18-39.

YAN S, XU D, YANG Q, et al. Discriminant analysis with tensor representation [C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005. Piscataway, NJ, USA: IEEE, 2005: 526-532.

TAO D, LI X, WU X, et al. General tensor discriminant analysis and Gabor features for gait recognition [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(10): 1700-1715.

LU H, PLATANIOTIS K N, VENETSANOPOULOS A N. Uncorrelated multilinear principal component analysis for unsupervised multilinear subspace learning [J]. IEEE Transactions on Neural Networks, 2009, 20(11): 1820-1836.

LU H, PLATANIOTIS K N, VENETSANOPOULOS A N. Uncorrelated multilinear discriminant analysis with regularization and aggregation for tensor object recognition [J]. IEEE Transactions on Neural Networks, 2009, 20(1): 103-123.

WANG S, YANG J, ZHANG N, et al. Tensor discriminant color space for face recognition [J]. IEEE Transactions on Image Processing, 2011, 20(9): 2490 -2501.

LIU Y, LIU Y, CHAN K C C. Tensor distance based multilinear locality-preserved maximum information embedding [J]. IEEE Transactions on Neural Networks, 2010, 21(11): 1848-1854.

JENSSEN R. Mean vector component analysis for visualization and clustering of nonnegative data [J]. IEEE Transactions on Neural Networks and Learning Systems, 2013, 24(10): 1553-1564.

SARKAR S, PHILLIPS P, LIU Z, et al. The human ID gait challenge problem: data sets, performance and analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 162-177.

HOFMANN M, GEIGER J, BACHMANN S, et al. The TUM gait from audio, image and depth(GAID)database: multimodal recognition of subjects and traits [J]. Journal of Visual Communication and Image Representation, 2014, 25(1): 195-206.

BEN X, MENG W, YAN R, et al. Kernel coupled distance metric learning for gait recognition and face recognition [J]. Neurocomputing, 2013, 120: 577-589.

BEN X, MENG W, YAN R, et al. An improved biometrics technique based on metric learning approach [J]. Neurocomputing, 2012, 97: 44-51.

彭亚丽,刘侍刚,裘国永.一种线性迭代非刚体射影重建方法.2015,49(1):102-106.[doi:10.7652/xjtuxb201501017]

周远,周玉生,刘权,等.一种适用于图像拼接的DSIFT算法研究.2015,49(9):84-90.[doi:10.7652/xjtuxb201509015]

毛彦斌,张选平,杨晓刚.伪DNA密码图像加密算法研究.2015,49(9):91-98.[doi:10.7652/xjtuxb201509016]

刘凯,张立民,孙永威,等.利用深度玻尔兹曼机与典型相关分析的自动图像标注算法.2015,49(6):33-38.[doi:10.7652/xjtuxb201506006]

吴一全,孟天亮,吴诗婳.人工蜂群优化的非下采样Shearlet域引导滤波图像增强.2015,49(6):39-45.[doi:10.7652/xjtuxb201506007]

唐晓庆,范赐恩,刘鑫.基于边缘保持滤波的单幅图像快速去雾.2015,49(3):143-150.[doi:10.7652/xjtuxb201503022]

杨晔,潘希德,庄健.一种针对超声检测图像的自适应阈值设置方法.2015,49(1):127-132.[doi:10.7652/xjtuxb201501 021]

侯兴松,张兰.方向提升小波变换域稀疏滤波的自然图像贝叶斯压缩感知.2014,48(10):15-21.[doi:10.7652/xjtuxb 201410003]

李博,冯松,毛军红.在线图像可视铁谱LED阵列反射光源设计与实现.2014,48(10):29-34.[doi:10.7652/xjtuxb2014 10005]

袁飞,朱利,张磊.利用超图图割的图像共分割算法.2014,48(02):20-24.[doi:10.7652/xjtuxb201402004]

符均,牟轩沁,季文博.亮色分离的饱和图像校正方法.2014,48(10):101-107.[doi:10.7652/xjtuxb201410016]

储颖,牟轩沁,洪伟.采用形状一致性特征的盲图像质量评价方法.2014,48(8):12-17.[doi:10.7652/xjtuxb201408003]

夏虎,庄健,于德弘.面向高维特征故障数据的进化软子空间聚类算法.2013,47(5):115-120.[doi:10.7652/xjtuxb201305 021]

刘剑毅,何苗,马瑶,等.面向人脸光照合成的光照参数精确估计算法.2013,47(6):18-24.[doi:10.7652/xjtuxb201306 004]

李玉花,齐春.利用位置字典对的人脸图像超分辨率方法.2012,46(6):7-11.[doi:10.7652/xjtuxb201206002]

蒋舒卉,钱学明.一种重要区域重采样的人脸检索方法.2012,46(2):119-123.[doi:10.7652/xjtuxb201202020]

岳桂华,滕奇志,何小海,等.岩心三维图像修复算法.2014,48(9):37-42.[doi:10.7652/xjtuxb201409007]

海涛,丛爽,鲍远律.彩色城市栅格地图道路网络自动获取方法.2011,45(12):16-21.[doi:10.7652/xjtuxb201112004]

罗涛,牟轩沁.一种胸部X射线摄影图像中结节检测的多尺度匹配滤波器.2011,45(4):30-35.[doi:10.7652/xjtuxb 201104006]

杨宏晖,戴健,孙进才,等.用于水声目标识别的自适应免疫特征选择算法.2011,45(12):28-32.[doi:10.7652/xjtuxb 201112006]

赵海峰,姚丽莎,罗斌.改进的人工鱼群算法和Powell法结合的医学图像配准.2011,45(4):46-52.[doi:10.7652/xjtuxb201104009]

余旺盛,侯志强,王朝英.频谱包络滤波器及其应用.2010,44(8):48-52.[doi:10.7652/xjtuxb201008010]

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Gait Recognition with Wavelet Moments of Silhouette Change Images

WiNet: a Gait Recognition Model Suitable for Wireless Sensing Scene

Energy Efficiency Optimization of the Cryogenic Air Separation Process for Electronic-Grade Ultra Pure Nitrogen

Review of the Application of EBSILON Software in Combined Heat and Power System Optimization

Diffusion Characteristics and Influencing Factors in Case of Hydrogen Leakage from Fuel Cell Vehicles Running in Tunnels

Related Author

陈实 1

高有行 1

段鹏松 1

周志一 1

王超 1

曹仰杰 1

王恩东 3

Related Institution

Institute of Computer Peripheral, Xidian University

Department of Design, Zhejiang Wanli University

School of Software Engineering, Zhengzhou University

⁰