核稀疏保留投影及在步态识别中的应用

王科俊; 阎涛; 吕卓纹; 唐墨

doi:10.11834/jig.20130302

图像分析和识别 | 浏览量 : 0 下载量: 271 CSCD: 0

PDF
导出
分享
收藏
专辑

核稀疏保留投影及在步态识别中的应用
Kernel sparsity preserving projections and its application to gait recognition
2013年18卷第3期页码：257-263
纸质出版：2013
DOI： 10.11834/jig.20130302
稿件说明：

移动端阅览

王科俊, 阎涛, 吕卓纹, 唐墨. 核稀疏保留投影及在步态识别中的应用[J]. 中国图象图形学报, 2013,18(3):257-263. DOI： 10.11834/jig.20130302.

Wang Kejun, Yan Tao, Lü Zhuowen, Tang Mo. Kernel sparsity preserving projections and its application to gait recognition[J]. Journal of Image and Graphics, 2013, 18(3): 257-263. DOI： 10.11834/jig.20130302.

摘要

针对已有基于线性变换的稀疏保留投影方法在解决实际问题时

会遇到维数灾难和小样本问题。通过引入核方法

提出一种核稀疏保留投影方法。首先采用非线性变换将原始数据映射到高维特征空间

而后在这个高维空间进行稀疏重构

并对得到的系数矩阵进行降维优化

最终得到所需的投影矩阵。将其应用到步态识别中

采用CASIA(B)步态数据库进行实验分析

实验结果表明

本文方法取得了令人满意的识别效果。

Abstract

In order to solve the problem of the curse of dimensionality and the small sample problem

a kernel sparsity preserving projection is proposed. First

the nonlinear transformation is used to map the original data to a high-dimensional feature space. Then

the sparsity reconstruction in a high-dimensional space is used and

the coefficient matrix is reduced and optimized. Finally

the projection matrix is obtained. This method is evaluated on the CASIA (B) Gait database. The experimental results show that the proposed method can obtain stable classification and performs satisfactory recognition results.