Facial Emotion Recognition Based on Viola-Jones Algorithm in the Learning Environment

An emotion is a trigger of learning success, so the learning should be adapting to the students' emotions. The most of popular approach is the acquisition of facial-based features. Therefore, we present facial emotion recognition based on the Viola-Jones Algorithm in the learning environment. Basically, the Viola-Jones algorithm is a face detection algorithm. However, we use facial-based features to detect face and recognize emotion, thus we applied rectangular feature and cascading AdaBoost algorithm which are the main concept of the Viola-Jones Algorithm in those both of process. In this study, we compare accuracy, precision, recall, and time-consuming of the Viola-Jones algorithm and our previous methods [1] using 50 UM's learning images in student emotion recognition. The accuracy, precision, recall, and time-consuming of Viola-Jones algorithm reach 0.74, 0.73, 0.76 and 15 seconds per frame, whereas our previous methods [1] reach 0.46, 0.48, 0.52, and 42 seconds per frame. In emotional recognition, we can conclude that the viola jones algorithm is superior to our previous research.