Smile Recognition Based on Support Vector Machine and Local Binary Pattern

In this paper, a method combining Local Binary Pattern(LBP) and Support Vector Machine(SVM) for smile detection is proposed. The process of smile recognition is divided into 5 parts including input images, image enforcement, face detection, feature extraction and classification. Firstly, the face images are downloaded from the Japanese Female Facial Expression(JAFFE) Database, which is then followed by the process of the image enforcement and processing such as noise removing and image normalization. After this, the human face extraction algorithm based on the combination of Haar features and cascading AdaBoost algorithm is used to segment the human face from the images. Furthermore, Local Binary Pattern(LBP) is applied to extract features from face images. Finally, Support Vector Machine based on Sequential Minimum Optimization(SMO) algorithm is implemented to classify the input feature vectors into two categories smiling images or not smiling images. The result shows that this method can get the accuracy of 88.1%.