On Face Recognition Using Hierarchical Self-Organized Gabor Features

Gabor-based face representation has achieve enormous success in face recognition. However, one drawback of Gabor-based face representation is the huge amount of data that must be stored. Due to the nonlinear structure of the data obtained from Gabor response, classical linear projection methods like principal component analysis failed to reduce this large amount of data. As a way to solve this problem, a nonlinear projection method is exploited. A set of hierarchical self-organizing maps is employed to capture the nonlinearity of the data and to represent it in a new reduced feature space. Experimental results on ORL face database prove the validity of our proposed feature ex-