LDA algorithm and its application to face recognition
Xie Yong-lin · Computer Engineering and Applications Journal · 2010
Linear Discriminant Analysis(LDA) is a linear feature extracting method using Fisher's criterion in face recognition.The main challenge of LDA is that its transformation matrix can not be computed directly;consequently,the discriminant vectors become very complicated,for the cases when the number of training samples is less than the dimensionality of the feature space.In this paper,a new LDA method is proposed using a modified Fisher's criterion.Experiments are conducted using ORL face database to present the superior performance of the proposed method,particularly for the cases when the number of training samples is small.