A Vision-Based Gesture Recognition and Student Engagement Assessment Model for Interactive Educational Environments

In the context of educational digitalization, traditional classroom methods for assessing student engagement-primarily based on teachers' subjective observations-suffer from limited real-time capabilities and lack of objectivity, making it difficult to accurately capture students' true interactive behavior.Taking management classrooms as an example, student actions such as raising hands or gesturing are key indicators of learning enthusiasm.Visionbased gesture recognition technology offers a novel, objective approach to engagement assessment.However, existing methods relying on depth cameras face challenges such as high equipment costs, poor environmental adaptability, and simplistic gesture-counting models that ignore contextual semantic information, making them unsuitable for complex classroom scenarios.This study focuses on interactive educational settings and proposes a gesture recognition and engagement assessment model based on image processing.First, we optimize image preprocessing, feature extraction, and pattern recognition algorithms for standard classroom environments to achieve high-precision, real-time recognition of various gestures such as hand-raising and waving.Second, by integrating multi-dimensional datagesture types, frequency, and duration-with instructional context, we construct a dynamic evaluation model that addresses the robustness issues of traditional methods in complex settings.The proposed approach offers a contactless solution for engagement assessment in smart classrooms, supports teachers in refining instructional strategies, and facilitates the digital transformation of educational interaction.This work holds significant implications for improving teaching quality and advancing educational technology innovation.