An improved age invariant face recognition using data augmentation

In spite of the significant advancement in face recognition expertise, accurately recognizing same individual across different ages still remains an open research question. Face aging causes intra-subject variations (such as geometric changes during childhood adolescence, wrinkles and saggy skin old age) which negatively affects accuracy systems. Over years, researchers have devised techniques to improve age invariant (AIFR) this paper, gesture network (FG-NET) dataset was adopted enable benchmarking experimental results. The FG-Net augmented by adding four types noises at preprocessing phase order trait features extraction training model used classification stages, thus addressing problem few available for dataset. developed adaptation a pre-trained convolution neural architecture (Inception-ResNet-v2) is very robust noise. proposed on testing achieved 99.94% accuracy, mean square error 0.0158 absolute 0.0637. results obtained are improvements comparison with related works.