Classroom Emotion Monitoring Based on Image Processing

Cèlia Llurba,Gabriela Fretes,Ramon Palau

DOI: https://doi.org/10.3390/su16020916

IF: 3.9

2024-01-23

Sustainability

Abstract:One challenge of teaching and learning the lack of information during these processes, including information about students' emotions. Emotions play a role in learning and processing information, impacting accurate comprehension. Furthermore, emotions affect students' academic engagement and performance. Consideration of students' emotions, and therefore their well-being, contributes to building a more sustainable society. A new way of obtaining such information is by monitoring students' facial emotions. Accordingly, the purpose of this study was to explore whether the use of such advanced technologies can assist the teaching–learning process while ensuring the emotional well-being of secondary school students. A model of Emotional Recognition (ER) was designed for use in a classroom. The model employs a custom code, recorded videos, and images to identify faces, follow action units (AUs), and classify the students' emotions displayed on screen. We then analysed the classified emotions according to the academic year, subject, and moment in the lesson. The results revealed a range of emotions in the classroom, both pleasant and unpleasant. We observed significant variations in the presence of certain emotions based on the beginning or end of the class, subject, and academic year, although no clear patterns emerged. Our discussion focuses on the relationship between emotions, academic performance, and sustainability. We recommend that future research prioritise the study of how teachers can use ER-based tools to improve both the well-being and performance of students.

environmental sciences,environmental studies,green & sustainable science & technology

What problem does this paper attempt to address?

The paper attempts to address the issue of the lack of collection and utilization of students' emotional information during the teaching process. Specifically, the authors believe that students' emotions play an important role in learning and information processing, and affect accurate understanding. Additionally, emotions influence students' academic engagement and performance. Therefore, considering students' emotions and their well-being helps to build a more sustainable society.

To achieve this goal, the study designed an emotion recognition (ER) model based on image processing to monitor students' facial expressions in the classroom. The model identifies faces, tracks action units (AUs), and classifies the emotions displayed by students through custom code, recorded videos, and images. The study further analyzed the relationship between these emotions and the academic year, subject, and classroom time points.

The results of the study indicate that there are various emotions present in the classroom, including both positive and negative emotions. There are significant differences in the occurrence of certain emotions based on factors such as the beginning or end of the class, subject, and academic year, but no clear patterns were observed. The discussion section focuses on the relationship between emotions, academic performance, and sustainable development, and suggests that future research should prioritize how teachers can use ER-based technological tools to improve students' well-being and performance.