Emotion-driven Activity Detection: Detecting Activities Perception through Facial Feature Analysis in Educational Environments

Understanding and categorizing student activities within educational settings is a critical component of enhancing learning outcomes and well-being. This study introduces an innovative approach to student activity classification by leveraging facial emotion recognition technology. The system is designed to capture and analyze students’ facial expressions, interpret their emotional states, and subsequently categorize their activities. By deploying a dataset encompassing a wide range of academic and non - academic activities, this research explores the purpose of deep learning models for recognition of facial emotion. The system is capable of identifying emotions such as sadness, surprise, happiness and anger. Extracted emotion features are then utilized to classify student activities, shedding light on whether a student is engaged, attentive, confused, or disinterested, among other states. This approach holds the potential to enhance educational environments by providing real-time insights into student behaviour and facilitating timely interventions to improve learning experiences and outcomes. Additionally, it opens up avenues for personalized educational support and the development of intelligent learning systems. In this paper we could implement the system to extract the facial features using Grassmann algorithm. And identify the emotions of student at predefined time intervals. Based on emotion classification, predict the active status and send reports to admin. Furthermore, this technology holds promise for the development of intelligent learning systems that adapt to students’ emotional states, providing additional support or challenges when needed. For instance, a virtual tutor could adjust the difficulty of exercises based on a student’s emotional responses, creating a dynamic and responsive learning experience.