Smart-Vision: Automated System for Attendance Management and Uniform Detection

This paper presents an automated attendance tracking system that integrates facial recognition with uniform detection to enhance reliability and efficiency in classroom attendance management. The system captures live video streams within a classroom environment and identifies students by matching detected faces with a pre-enrolled database using a deep learning–based face recognition framework. Simultaneously, a YOLOv8-based object detection model verifies whether the detected individual is wearing the prescribed uniform. Attendance is recorded only when both identity verification and uniform compliance are successfully confirmed. The system performs real-time processing using OpenCV with a Flask-based backend architecture, while a React.js web interface provides an intuitive platform for monitoring attendance records and system outputs. Experimental evaluations conducted in a real classroom environment demonstrate strong performance, achieving over 95% accuracy for face recognition and more than 90% accuracy for uniform detection. The proposed framework effectively handles challenges such as variations in lighting conditions, occlusions, and similarities in clothing patterns, thereby improving the reliability, security, and efficiency of automated attendance monitoring systems.