PCA applied to neural internal representation of input data - application to face recognition

M. Oravec1
1Department of Telecommunications, Faculty of Electrical Engineering and Information Technology, Slovak University of Technology, Bratislava, Slovakia

Tóm tắt

An original method of feature extraction from image data is introduced using MLP (multilayer perceptron) and PCA (principal component analysis). This method is used in a human face recognition system and results are compared to a face recognition system using PCA directly and to a system with direct classification of input images by MLP.

Từ khóa

#Principal component analysis #Face recognition #Multilayer perceptrons #Eigenvalues and eigenfunctions #Neurons #Feature extraction #Image databases #Testing #Neural networks #Information technology

Tài liệu tham khảo

pavlovicova, 2001, Generalized Interpolation of Segmented Images Using Shape-Independent Orthogonal Transforms, Proc of Int Conference on Image Processing ICIP 2001, 888 10.1162/jocn.1991.3.1.71 haykin, 1994, Neural Networks - A Comprehensive Foundation Macmillan College Publishing Company