Detection of range and degrees of emotion through facial Image analysis using Machine Learning for understanding students’ feelings

Abstract

This work concentrates on two issues. Firstly, to develop an emotion detection system based on Machine Learning algorithms applied to facial and allied analyses, in particular, detecting the dilation features of the pupil of the eye, so that various degrees of emotions and attentiveness can be detected. This would answer questions like, in a classroom teaching environment, how attentive is a student – that is to what degree or percentage – in a class, how happy, sad or scared – once again to what degree or percentage – a person is in response to an event or in an environment, and questions like that. Secondly, the system should be able to detect these emotions as displayed by a group of people, and not just an individual. As it is seen in existing literature, facial analysis research has concentrated on many aspects like social norms, refinement of analysis, reduction of noise etc. However, the problem that this research is addressing has dealt with in very fewinstances of research. It may thus be concluded that inthe issue of mass scale emotion detection, a lot of research is yet to come. It is hoped that the findings of this research will pave way to solving problems like understanding student behaviour in a teaching environment, understanding crowd behaviour in an open environment, understanding audience emotions in response to a piece of art, and so on.

DOI - https://doi.org/10.65525/SVUP.9788199963351.2026.67-86