Age Estimation Using Classifiers Artificial Neural Network and Support Vector Machine Based on Face Images

The most prominent challenge in the facial age estimation is a lack of sufficient and incomplete training data. Aging is slower and gradual process, therefore, faces near close ages look quite similar this can allow us to utilize the face images at neighbouring ages with modeling to a particular age. There are many potential applications in age-specific human-computer interaction for security control and surveillance monitoring. In the last few years biologically inspired features are used for human age estimation for face images but recently more focus put on a method like scattering transform. The proposed approach exploits scattering transform gives more information about features of the facial images. An efficient descriptor consisting scattering transform which scatters the gabor coefficients and pulled with Gaussian smoothing in multiple layer and is evaluated for facial feature extraction. These extracted features are classified using support vector machine and artificial neural network. Results for face based age estimation obtain by the artificial neural network is more effective than support vector machine.