Research On Facial Expression Recognition Algorithm Based On Static Images

With the advent of the information age, computer has quietly changed the way people live, become a part of people’s daily study, work and life. In the process of frequent use of computers, people are increasingly looking forward to a friendlier human-computer interaction experience. Facial expression, as a part of human biological characteristics, reflects the complex and subtle emotional changes in people’s hearts, and conveys a lot of emotional information. If computers can understand their inner feelings according to the user’s facial expression, realize the recognition of facial expression, people can experience more efficient, more user-friendly human-computer interaction, at the same time, facial expression recognition technology can also be used in many fields of social life, such as intelligent human-computer interaction, medical, intelligent security monitoring and emotional state analysis.Therefore, the study of facial expression recognition has a very important role in improving people’s life quality and providing people with convenient and quick lifestyle. This paper is based on the static images of the face recognition algorithm research, the research object is the static facial expression image, the main research content is mainly related to facial expression feature extraction method, effective feature selection method and the design and implementation of facial expression classification algorithm. In this paper, facial expression recognition algorithm based on static image has been researched, the research object is the static facial expression image, and the main research content is mainly related to the design and implementation of facial expression feature extraction method, effective feature selection method and facial expression classification algorithm.In this paper, in order to overcome the shortcoming of data redundancy of the traditional Gabor feature, sparse facial expression recognition algorithm based on integrated Gabor feature is proposed. Firstly, by means of two integration methods, mean fusion and differential binary encoding, the original Gabor feature images are integrated in a multi-scale and multi-angle way and 26 integrated Gabor feature images are obtained; then use feature selection method based on the facial expression recognition contribution coefficient, selecting 4 images from 26 integrated Gabor feature images as the final feature vector. Finally, the over complete dictionary of sparse representation classifier is constructed with the final feature vectors of the training samples, sparse representation on images from the final feature vectors corresponding to facial expression image is solved subject to an L1 objective function. All the reconstruction errors between the test sample and the reconstructed sample of each type of expression are calculated, and the expression corresponding to the minimum reconstruction error is selected as the final expression recognition result. Experimental results indicate that the proposed sparse facial expression recognition algorithm based on integrated Gabor feature, can separate and express the facial expression features facial features effectively, and reduce dimension and present expression data compactly; meanwhile the expressions are classified correctly.