《计算机应用研究》|Application Research of Computers

基于<i>k</i>-最小表示误差类的表示分类方法

Representation-based classification based on <i>k</i>-minimum representation error classes

免费全文下载 (已被下载 次)  
获取PDF全文
作者 罗智玉,郑成勇
机构 五邑大学 数学与计算科学学院,广东 江门 529000
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2021)10-026-3035-05
DOI 10.19734/j.issn.1001-3695.2021.01.0049
摘要 基于表示的分类(representation-based classification,RC)通常使用所有类的训练样本来表示测试样本。然而,是否需要使用全部类来表示测试样本仍有待研究。为此,提出一种两阶段表示分类框架。首先使用RC算法计算测试样本相对于全部类的训练样本的表示系数,找出前<i>k(k≥1)个具有最小表示误差的类;然后利用该k</i>个类的训练样本,再次应用RC算法对测试样本进行表示,并通过从这<i>k</i>个类中找出最小表示误差类来确定测试样本的类别。此外,提出了一种非负加权协同表示分类算法。所提分类框架中的前后两个RC算法可以相同也可以不同。取前后两个RC相同,对五种RC,在五个数据库上进行实验,实验结果表明,所提两阶段表示分类框架大多数情况下能显著提升原RC算法的分类精度。
关键词 基于表示的分类; <;i>;k<;/i>;-最小表示误差类; 两阶段; 非负加权; 协同表示
基金项目 广东省自然科学基金资助项目(2018A030313063)
本文URL http://www.arocmag.com/article/01-2021-10-026.html
英文标题 Representation-based classification based on <i>k</i>-minimum representation error classes
作者英文名 Luo Zhiyu, Zheng Chengyong
机构英文名 School of Mathematics & Computer Science,Wuyi University,Jiangmen Guangdong 529000,China
英文摘要 Representation-based classification(RC) usually uses all the training samples of all pattern classes to represent a test sample. However, whether it is necessary to use all pattern classes to represent the test sample remains to be investigated. So, this paper proposed two-stage representation-based classification framework. First, it used a RC algorithm to calculate the representation coefficients of the test sample w. r. t the training samples of all classes, and found the first <i>k(k≥1</i>) classes with the minimum representation error. Then with the training samples of these <i>k</i> classes, it applied a RC algorithm again to represent the test sample, based on which, the label of the test sample was finally determined by finding the minimal representation error class among those <i>k</i> classes. In addition, this paper developed a nonnegative weighted collaborative RC(NWCRC) algorithm. The first and second RC algorithm of the proposed RC framework could be the same or different. Setting the first and second RC algorithms to be the same, with five different RC, experiments were carried out on five databases, experimental results of which show that the proposed two-stage RC framework can significantly improve the classification accuracy of the original RC algorithms in most cases.
英文关键词 representation-based classification; < i> k< /i> -minimum representation error classes; two-stage; nonnegative weighted; collaborative representation
参考文献 查看稿件参考文献
 
收稿日期 2021/1/27
修回日期 2021/3/2
页码 3035-3039
中图分类号 O235;TP391.4
文献标志码 A