Development of a facial expression recognition system for the laughter therapy

This paper proposes a facial expression recognition system for the laughter therapy. The proposed system takes two steps: face detection and facial expression recognition. At the face detection stage, candidate facial areas are detected in real time from images taken by a camera in consideration of Haar-like features, followed by the application of a SVM(Support Vector Machine) classifier to detect face images in a more correct way. Next, histogram matching-based illumination normalization is used to mitigate the influence of lighting on the detected images. At the facial expression recognition stage, PCA (Principle Component Analysis) is used to capture features of the face, and real-time laugher recognition is made via a multi-layer perceptron artificial neural network. From the findings of this study, we conclude that the proposed method can improve facial expression recognition through illumination normalization based on histogram matching and by testing candidate facial images with a SVM.