Decoding Student Emotions: An Advanced CNN Approach for Behavior Analysis Application Using Uniform Local Binary Pattern

In this paper, Facial Emotion detection model, built using Uniform Local Binary Pattern, is explained using an application i.e., Student behavior detection. Student engagement is a critical factor in effective teaching and learning. Conventional approaches to evaluating student participation frequently depend on assertions made by students and observations made by teachers, which can be laborious and unreliable. The use of computer vision techniques to automatically identify the emotions and behaviors of students in real time has gained popularity in recent years. This research provides a novel computer vision-based method for detecting student behavior, emphasizing emotion recognition. We propose a system that utilizes Convolutional Neural Networks (CNNs) for emotion classification, data augmentation to enhance dataset, and Uniform Local Binary pattern (uLBP) feature extraction for improved texture analysis. The system aims to provide educators with real-time insights into student emotions, enabling them to adapt their teaching strategies accordingly. In order to expand the variety and breadth of the training dataset, we enhanced the facial expression data that we extracted from a dataset that’s well suited for this work. We then applied uLBP feature extraction to capture local texture patterns in the facial images, which were then used as input to the CNN for emotion classification. According to our findings, the suggested approach successfully classified student emotions with a high accuracy of 94% and also completed the ablation study, which indicates its usefulness in practical settings.