Analysis of classroom learning behaviors based on internet of things and image processing

Abstract

The quick and accurate identification of classroom emotions helps teachers perceive the learning state of their students. This paper designs a bimodal identification system for classroom emotions based on electroencephalogram (EEG) signals and countenances. The system relies on the Internet of things (IoT) technology to collect EEG signals, and extracts the signal features with fractal dimension and multiscale entropy algorithm. After that, the support vector machine (SVM) was adopted to classify the classroom emotions. Then, the features of countenances were extracted by local binary pattern (LBP). Experimental results show that our system accurately identified 85.7% of classroom emotions. Compared with the traditional countenance-based emotion identification method, the bimodal approach could extract rich information on classroom emotions, and achieve a good effect on emotion identification.