摘要
人脸自动识别是模式识别领域中一项具有广阔应用前景和实际应用价值的热门课题 .文中提出了一种主元分析和神经网络相结合的方法进行人脸识别 .我们先对图像进行归一化处理 ,采用主元分析法对图像的主特征分量进行提取 ,然后以一个径向基函数神经网络作为分类器 ,进行人脸识别 .实验结果表明 ,这种方法在采用多样本训练后 ,具有较高的识别率 .
The automatic recognition of human faces, which has a wide range of potential applications, is an active research area of pattern recognition. In this paper, an algorithm combining principal component analysis and neural networks is proposed. Firstly, principal component analysis is used to extract the principal features of unitary face images. Then a radial basis function (RBF) neural network is designed as a face classifier to perform recognition. Simulations show that the method can reach a high recognition rate after multi-sample training.
出处
《山东大学学报（工学版）》
CAS
2004年第2期55-58,共4页
Journal of Shandong University（Engineering Science）
关键词
主元分析
人脸识别
神经网络
模式识别
principal component analysis
face recognition
neural networks
pattern recognition
