摘要
对在不同视角下,得到的人脸模型,文中提出一种基于人脸表面的识别方法,采用平面射影变换,将人脸的图像变换到一个相同的图像,使图像对齐;而后运用主成分分析法(PCA)进行分类。基于这种方法,由光线、面部表情、姿态的改变引起的不必要变化就可能被消除或可以忽略。这种方法可以达到比较准确的识别人脸的目的。实验结果显示,文中的方法对人脸模型提供了更好的表达,并且人脸识别的错误率更低。
In this paper,a novel method is proposed to recognize faces in different pose. First, face images are changed into another same face image by the planer projective transformation, then they are represented and classified by the principal component analysis（PCA）. By this way,the unwanted variations resulting from changes in lighting, facial expression,end pose may he eliminated or reduced. The valuations on the actual images have shown that our methods can provide a better representation and achieve lower rates in face recognition.
出处
《计算机技术与发展》
2006年第6期213-215,218,共4页
Computer Technology and Development
关键词
主成分分析
线性分类分析
射影变换
principal component analysis（PCA）
linear classification analysis
projective transformation
