基于CCA的增强线性鉴别分析

为了有效地融合Fisher线性鉴别分析与最大散度差鉴别分析所抽取的特征,得到更加全面反映原始样本的鉴别特征集,提出了基于典型相关分析的增强线性鉴别分析方法.利用Fisher线性鉴别分析和最大散度差鉴别分析方法提取两组鉴别特征,根据典型相关分析对这两组特征进行融合,获得更具鉴别力的典型鉴别特征.经过ORL标准人脸库实验,验证了所提算法的有效性.

Enhanced Linear Discriminant Analysis Based on Canonical Correlation Analysis

To effectively mix the features which are extracted by Fisher linear discrininant analysis and maximum scatter difference discriminant analysis and form a feature set which can reflect the samples more comprehensive,an enhanced discriminant analysis method based on canonical correlation analysis is proposed in the paper.Fisher linear discriminant analysis(LDA) and Maximum Scatter Difference Discriminate Analysis(MSDDA) are firstly adopted to extract two sets of features in the same pattern space,respectively.The canonical correlation analysis method is then used to fuse the two sets of features obtained above and derives more effective canonical discriminant features.Finally,the extensive experiments are performed on ORL face database.The experimental results verify the effectiveness of the proposed method.