Automated Emotion-Driven Attendance System: A CNN-Based Facial Recognition Platform with Real-Time Emotion Detection and System-Level Validation

Neelamsetti Preethi¹, Penta Shobitha¹, Yedhuru Ganesh¹, Payyapalli Hymavathir¹, P. Ramya²

¹^,2Department of Information Technology, MVGR College of Engineering (A), Vizianagaram, India

Abstract: Emotion recognition and automated attendance tracking have emerged as significant applications of artificial intelligence in modern educational and organizational environments. Traditional attendance systems are often manual, time-consuming, and prone to inaccuracies, while lacking the capability to capture user engagement or behavioral context. This paper presents a deep learning-based framework for an automated emotion-driven attendance system capable of performing real-time facial recognition and emotion detection, eliminating the need for manual intervention.The proposed system is built using Convolutional Neural Networks (CNN) for robust facial recognition and leverages advanced emotion detection techniques to classify human expressions into multiple emotional states such as happy, sad, neutral, and angry. The model processes live video input, detects faces using computer vision methods, and matches them against a pre-trained dataset for identity recognition. Upon successful identification, attendance is automatically recorded along with the corresponding emotional state, ensuring both efficiency and contextual awareness.

To enhance system reliability, the framework is designed to operate under real-time conditions with optimized preprocessing techniques, including image normalization and feature extraction. The system demonstrates strong performance in face recognition and emotion classification, achieving consistent accuracy across standard testing conditions, with minor limitations observed under low-light environments or partial occlusions. In addition to attendance marking, the system provides a structured record of detected emotions, enabling improved monitoring of user engagement.Overall, the proposed system functions not only as an automated attendance solution but also as an intelligent monitoring platform, combining identity recognition and emotional analysis within a unified architecture. This makes it suitable for deployment in classrooms, workplaces, and other real-world environments requiring efficient and adaptive attendance management.

Keywords— Emotion Detection; Facial Recognition; Deep Learning; Convolutional Neural Networks (CNN); Automated Attendance System; Real-Time Processing; Computer Vision; Human Emotion Recognition