A novel age estimation of facial image using KLT

In this paper, a proficient feature extraction method has been developed which is based on the Kande-Lucas-Tomasi (KLT) algorithm. Each face image is extracted and is marked for Anthropometric Model (AM) distance measure. The KLT facial coefficients of dissimilar scales & a range of angles are obtained. The coefficients are in use as a feature vector for advance processing. The Fast-Independent component analysis (ICA) algorithm is adopted on face image based on entropy to take out the face characteristic information. Finally, according to the anthropometric model to categorize face feature and artificial neural network (ANN) used to calculate age for all kind of facial databases. Experiments are carried out using the YALE database. An investigational result shows that the recognition rate and Mean Absolute Error (MAE) of the projected algorithm is satisfactory and extremely promising.