A Novel Model for Independent RBF Neural Networks Employing Gabor-based Kernel PCA with Fractional Power Polynomial Models for Feature Extraction

A novel model for independent radial basis function (IRBF) neural network employing Gabor-based kernel PCA with fractional power polynomial models for feature extraction is proposed in this paper. In the new model, a bank of Gabor filters is first built to extract Gabor face representations characterized by selected frequency, locality and orientation to cope with various illuminations, facial expression and poses in face recognition. After extracting Gabor face representations for every face sample, a kernel PCA with fractional power polynomial models is chosen to extract high-order statistical features of extracted Gabor face representations. At last, a new IRBF neural network is built to classify these extracted high-order statistical features of Gabor face representations. According to the experiments on the famous CAS-PEAL face database, our proposed approach could outperform PCA, ICA with architecture II (ICA2) and kernel PCA (KPCA) with standing testing sets proposed in the current release disk of the CAS-PEAL face database