Facial Expression Feature Extraction Based on FastLBP

The methods of facial expression feature extraction based on traditional LBP have some drawbacks such as complexity, high dimension of feature vectors, which may reduce the efficiency of subsequent recognition process. To solve the above problems, this paper proposes an improved LBP algorithm——FastLBP(FLBP). FLBP compresses the feature vectors described by LBP histogram to decrease the complexity of the algorithm. This method increases the efficiency of training and testing in facial expression recognition and ensures the accuracy of recognition at the same time. The experimental results show that FLBP proposed by this paper is fast and effective.