Feature Selection of Face Recognition Based on Improved Chaos Genetic Algorithm

Aiming at the problem of how to determine the dimensions of the eigenvectors in principal component analysis (PCA), this paper presents a novel feature selection method based on improved chaos genetic algorithm (ICGA). First, two kinds of chaotic mappings are introduced in different phase of ICGA, which maintain the diversity of population and enhance the global searching capability; Second, this paper make use of PCA to extract eigenvectors of the face images. Then, feature (eigenvector) selection using ICGA, which can quickly find out feature subspace that is most beneficial to classification. The experimental results based on ORL face database indicate that the proposed method not only reduces the dimensions of feature space, but also achieves higher recognition rate than other methods.