Design and Development of an Automated Facial Recognition System for Student Attendance Monitoring

Abstract

Traditional manual attendance methods in virtual classrooms are prone to human error and security vulnerabilities, necessitating automated solutions that prevent fraudulent proxy attendance. This research presents the development of a multimodal biometric tracking system implementing specific check-in and checkout logic to accurately record student participation in virtual environments. The study is grounded in the Unified Theory of Acceptance and Use of Technology and Diffusion of Innovations theory to evaluate the critical factors influencing student adoption of biometric systems. A quantitative survey design was employed with a sample of 903 valid questionnaires collected from college students in Wuhan, China. Data were analyzed using structural equation modeling with the Analysis of Moment Structures software. Performance expectancy predicted student acceptance intention most strongly (beta = 0.667, p < 0.01), while trust in the system also demonstrated a significant positive effect (beta = 0.183, p < 0.001). Perceived privacy risk was found to significantly reduce students' trust in the biometric protocol (beta = -0.475, p < 0.001). Notably, effort expectancy did not significantly associate with student intention to adopt the automated tracking system (beta = -0.082, p > 0.05). This work uniquely integrates a multimodal biometric architecture with a comprehensive multi-faceted risk perception model in a higher education setting. Educators and system designers should prioritize enhancing performance expectations and implementing robust privacy safeguards to improve student adoption of automated attendance tracking.