IMPLEMENTATION OF THE DATA UNCERTAINTY MODEL USING APPEARANCE BASED METHODS IN FACE RECOGNITION

The face images should not be the fully accurate to representation and for an observation. To reducing the uncertainty for representation of the face images and it is to improving the accuracy of face recognition, more observation of the same person face images is required in the face recognition. The data of face images are obtained from different pose, facial expression and, hence a single image of the face occurring the high uncertainty for the face representation. In this paper, develop such a model which is to improve the accuracy in the face recognition by reducing the data uncertainty. The model is to reduce the uncertainty of face images representation by synthesizing the virtual training samples. Here, propose system gives the useful training samples which are selected, and are comparable to the test sample from the set of all the original training samples and synthesized virtual training sample. The proposed work will give the classified virtual training samples which are not contained the duplication of the samples in the model. In the real world face recognition system the uncertainty highly occurred because the limited number of available face images of subject and due to this there is high uncertainty is occurred