Age Invariant Face Recognition Using Artificial Neural Network

Abstract

In current research, face recognition technology is being used to improve human efficiency when recognizing
faces, is one of the fastest growing fields in the biometric industry. The challenges in developing age invariant face
recognition are large intra-subject variations and large inter-user similarity. The main intra-subject variations are
(pose, illumination, expression, and aging) commonly encountered in face recognition. In this paper, feature extraction
based age invariant face recognition framework is modelled. In this system first component or features are extracted
using two local feature extraction SIFT (scale invariant feature extraction scheme) and MLBP (multi-scale local binary
pattern) methods. This feature vector is then classified using artificial neural network. These results are normalized and
using score sum fusion rule recognition in performs. Evaluation of this system is done by different measuring parameters
to check the robustness of the proposed system. Experimental results shows that the accuracy is increased up to 53%
compared to existing approach.