提出了一种基于核主成分特征组合的人脸识别方法。
A new face recognition method based on combination of KPCA features is proposed in this paper.
基于OR L人脸库，识别核主成分分析提取出的主成分的相关性系数。
Based on ORL face database, recognizes correlation coefficients of principal component extracted by KPCA.
基于多层核主成分提取估计器需要将调制信号的训练样本根据各自的频率进行分层。
The estimator based on kernel principal component extraction requires to stratify the training samples of interested signals with respect to their respective frequencies.
本文的主要工作是将支持向量机(SVM)及核主成分分析(KPCA)应用到入侵检测技术中。
The dissertation mainly aims at applying support vector machine (SVM) and kernel principal component analysis (KPCA) to intrusion detection.
通过实验比对可知该算法效果在特征提取和分类方面均优于传统核主成分分析法以及最近邻分类器。
The experimental comparisons show that this algorithm outperforms traditional KPCA and K-Nearest Neighbor classifier on both feature extraction and classification.
基于核主成分分析（KPCA）的人脸识别算法能够提取非线性图像特征，在小样本训练条件下有较好性能。
The algorithm of face recognition based on kernel principal component analysis(KPCA)can abstract nonlinear features of image and can get better performance under less sample training conditions.
论文介绍了基于核空间的ICA的原理和基本算法，然后介绍了该算法与典型ICA和主成分分析(PCA)在盲源信号分离中的比较。
In this paper, kernel independent component analysis (KICA) 's principle and algorithm are introduced, and then the KICA comparison with some other ICA and principal component analysis (PCA) is given.
结果表明，通过对核参数的适当选取，可使最大特征值的贡献率达到或接近85 %，避免了多个主成分的不同组合而导致的评价结果的不一致。
The results showed that the maximum eigenvalue contributes nearly 85 % by choosing appropriate parameters, avoiding the different array as a result of many principal composition.
结果表明，通过对核参数的适当选取，可使最大特征值的贡献率达到或接近85 %，避免了多个主成分的不同组合而导致的评价结果的不一致。
The results showed that the maximum eigenvalue contributes nearly 85 % by choosing appropriate parameters, avoiding the different array as a result of many principal composition.
应用推荐