The exploration of facial expression recognition in distance education learning system

Abstract

In recent years, the learning style of modern distance education has been more and more popular among the learners. However, the learner’s emotion is often ignored during the distance education learning process. In this paper, the study purpose is mainly concerned with how to effectively recognize the emotion through the way of using facial expression for the future distance education learning. We apply the method of Convolutional Neural Network (CNN) in our research. First, we introduce the structure of CNN in terms of the convolutional layers, sub-sampling layers and fully connected layers. Secondly, we propose a framework to use CNN in distance education system. Thirdly, we carry out experiment on a data set that consists of facial expression images of learners to evaluate the performance of proposed method. Finally, we get the conclusion the average accuracy of CNN to recognize the facial expression is 93.63%. The high accuracy shows the application of CNN to recognize facial expression is valuable and helpful for the teachers in the future distance education system to attain and understand the learners’ emotion state in real time, accordingly to regulate the teaching strategy in time.