基于范数加权的二维主成分分析提取图像特征

Two-dimensional principal component analysis based on norm weighted to extract image features

为了能准确地进行图像特征提取,提出一种新的基于Nuclear范数与Frobenius范数加权的二维主成分分析(记为NF-2DPCA)方法。并利用来自于Yalefaces、AR数据库以及ORL人脸数据库中的数据,进行了图像面部识别和重建等相关实验,结果表明所提出的方法能够有效地提取图像特征,并且面部识别准确率最高能达到94.25%,进一步显示所提方法具有一定优越性。

In order to extract image features accurately,a new Two-dimensional principal component analysis(NF-2DPCA)method weighted by Nuclear norm and Frobenius norm was proposed.Using the data from Yalefaces,AR database and ORL face database,the relevant experiments of image face recognition and reconstruction were carried out.The results showed that the proposed method can effectively extract image features,and the highest accuracy of face recognition can reach 94.25%,which further showed that the proposed method has certain advantages.