The Research Of Facial Expression Recognition Algorithm Based On Gabor Wavelet

The goals of expression recognition are to enable computer to recognize facial expression information automatically, and further enhance the friendliness and intelligence of man machine interaction. However, due to facial expression recognition involves image processing, computer vision, applied mathematics and other subjects, Because of this complexity and particularity, there are a lot of problems need to solve.A typical Facial Expression Recognition system usually contains the following three Parts:face detection, facial expression feature extraction and classification of facial expression.The main purpose of this paper is to recognize human facial expression. The research in this paper focused on the extraction and analysis of the facial features. This paper presents a new FER algorithm based on 2D Gabor transforms, Adaboost algorithm and SVM.Firstly, based on deeply researches on SVM and the Adaboost algorithm, this approach takes the advantages of the favorable ability of Gabor feature in representing expression variability, the effective function of Adaboost in feature selection, and the high performance of SVM in the solution to small sample size, high dimension problems.Secondly, since the high dimensional Gabor feature vectors are quite redundant, Adaboost is introduced as a method to reduce the dimensionality of feature vectors, and to extract Gabor features of the face image.Thirdly, SVM algorithm is based on statistical learning theory, and SVM algorithm has good Properties in dealing with nonlinear Problems, linear non-separable Problems and small samples problems. To solve the multi-class classification problem, we designed classifier by one-to-one mode, so the number of classifiers of SVM was k(k-1)12(k is the number of categories).Finally, this paper selects the JAFFE expression data base as the training sample, and verifies recognition result of this combined algorithm. This paper presents an algorithm for happy, surprise, anger, disgust four kinds of facial expression recognition results achieved an average 93.6% more than the previous results by 5 percentage points.