Face Recognition using PCA and LDA with Singular Value Decomposition(SVD) using 2DLDA
Neeta Nain, Prashant Gour, Nitish Agarwal, Rakesh P. Talawar, N. Subhash Chandra · 2008
Linear Discriminant Analysis(LDA) is well-known scheme for feature extraction and dimen- sion reduction. It has been used widely in many appli- cations involving high-dimensional data, such as face recognition. In this paper we present a new variant on Linear Discriminant Analysis (LDA) for face recogni- tion by reducing dimensions of input data using ma- trix representation and after that using singular value decomposition to reduce dimensions of scatter matrix. Experiments on ORL face database shows the effec- tiveness of our proposed algorithm and results com- pared with other LDA based methods shows that the proposed scheme gives comparatively better results than previous methods in terms of recognition rate and reduced time complexity.