摘要
提出了KPCA的一种称为联合核主成分分析(Joint Kernel Principle Component Analysis,JKPCA)的变型,能够从输入和输出空间引出先验信息用于特征提取.首次将联合核映射应用于特征提取领域,而且在图像数据集上的实验结果表明,JKPCA是可行并有效的.
A variant of KPCA called joint kernel principal component analysis（JKPCA）is proposed,which is able to obtain the prior information from both the input and output spaces for the feature extraction processing.Moreover,the joint kernel mapping is applied into feature extraction for the first time.The experimental results validate the feasibility and effectiveness of the proposed JKPCA on image datasets.
出处
《沈阳大学学报（自然科学版）》
CAS
2015年第4期306-312,共7页
Journal of Shenyang University：Natural Science
基金
国家自然科学基金资助项目(61272198)
关键词
核主成分分析(KPCA)
联合核映射
特征提取
核方法
kernel principal component analysis（KPCA）
joint kernel mapping
feature extraction
kernel-based method
