Feature representation and classification are two key steps for face recognition. We compared three automated methods for face recognition using different method for feature extraction: PCA (Principle Component Analysis), LDA (Linear Discriminate Analysis), ICA (Independent Component Analysis) and SVM (Support Vector Machine) were used for classification. The experiments were implemented on two face databases, The ATT Face Database [1] and the Indian Face Database (IFD) [2] with the combination of methods (PCA+ SVM), (ICA+SVM) and (LDA+SVM) showed that (LDA+SVM) method had a higher recognition rate than the other two methods for face recognition.

ABSTRACT

The face is the first source of information that inspires the attractiveness of a human being, for this reason; several studies were conducted in the aesthetic medicine or in the image processing to analyse the aesthetic quality of an adult human face. This paper proposes an automatic procedure for the analysis of facial beauty. First, we detect the face zone on an image and its features areas, then we present our novel method to extract features corners, and finally we analyse the facial aesthetic quality. Experimental results show that our method can extract the features corners accurately for the majority of faces presented in the European Conference on Visual Perception in Utrecht (ECVP) and Faculdade de Engenharia Industrial (FEI) images databases, and that there exist a difference in the facial beauty analysis by gender and age, due to anatomic differences in specific facial areas between the categories.

In the field of face recognition, the major challenge that encountered classification algorithms, is to deal with the high dimensionality of the space representing data faces.Many methods have been used to solve the issue, our focus, in this paper, is to compare the efficiency (in the term of complexity and recognition rate) of linear and non linear dimensionality reduction methods. We study the influence of high and low dimensionality of features using PCA, LDA, ICA and Sparse Random Projection.Experiments show that projecting the data onto a lower dimensional subspace using non linear method give a high face recognition rate.