Study On A Kind Of Cooperative Representation Face Recognition Algorithm

With the development of computer technology,face recognition technology is playing an important role in many fields.As we know,the main challenges of face recognition are that the face image might severely vary with the various poses,facial expression and illumination.In order to improve Collaborative Representation Classification algorithm,we consider this problem from different perspectives.The specific work done in this paper has the following aspects:1.A new method based on the combination of mirror image and coarse-to-fine face recognition is proposed in this paper.The new method firstly use the mirror image of the face image to generate new samples,and then devised representation based method simultaneously uses the original and new training samples to perform a sparse coarse-to-fine representation.2.A new collaborative representation classification method based on multiple images for face recognition is proposed in this paper.The new method firstly use the mirror feature of the face image to generate new samples,and then use the arbitrary two new samples and original samples respectively to synthesize the smoothing median virtual samples,we select useful training samples that are similar to the test samples from the set of all the original and synthesized virtual training samples based on the Euclidean distance.We design a parameter weighted fusion for training samples composed of different approaches respectively.3.A new method based on the new discriminative probabilistic collaborative representation for robust face recognition is proposed in this paper.The new method uses the objective function of the l2-norm collaborative representation classification based on the total probability form to generate a new objective function,which forms a new decision algorithm,it jointly maximizes the likelihood that a test sample belongs to the corresponding class.