Title :
A new algorithm of uncorrelated discriminant vectors and its application
Author :
Xiao-jun, Wu ; Shi-Tong, Wang ; Tong-Ming, Liu ; Jing-Yu, Yang
Author_Institution :
Dept. of Electron. & Inf., East China Shipbuilding Inst., Zhenjiang, China
Abstract :
A study has been made on the method of solving an optimal set of uncorrelated discriminant vectors. We have made an improvement to the method of solving an optimal set of uncorrelated discriminant vectors of Zhong Jin et al. (2001) on the basis of the equivalence of two Fisher criteria. A concise formula for solving the optimal set of uncorrelated discriminant vectors has been proposed. Numerical experiments on the facial database of ORL show the effectiveness of the proposed method.
Keywords :
covariance matrices; face recognition; feature extraction; pattern classification; statistical analysis; vectors; Fisher criteria; ORL; facial database; optimal set; uncorrelated discriminant vectors; Covariance matrix; Face recognition; Feature extraction; Humans; Image recognition; Pattern recognition; Principal component analysis; Scattering; Spatial databases; Vectors;
Conference_Titel :
Intelligent Control and Automation, 2002. Proceedings of the 4th World Congress on
Print_ISBN :
0-7803-7268-9
DOI :
10.1109/WCICA.2002.1022125