Abstract:
An efficient Illumination and Pose Invariant (IPI) methodology to recognize human faces under uncontrolled lighting and the age of the recognized faces is proposed in this paper. Face recognition is based on robust pre-processing followed by fusion of Extended Curvature Gabor wavelets (ECG) and Local Binary Patterns (LBP) to extract the features of curvature information and the texture information of face image respectively. As both feature sets are higher in dimension, PCA is used to reduce the dimensionality prior to Z-Score normalization. Nearest Neighbour classifier is used to recognize the face. Support Vector Machine based Regression algorithm is used to estimate the age of a face in an image. The proposed IPI method found to give better recognition accuracy than the existing methods available in the literature. The results of the proposed system are evaluated using Extended Yale-B database and our self collected MIT India database of different age groups.
