ABSTRACT: Attendance management is an important administrative activity in educational institutions and workplaces. Traditional methods such as calling names or signing paper registers require time, depend on manual entry, and may allow proxy attendance. This paper presents a framework for a Facial Recognition Attendance System that uses a camera, computer vision, and machine-learning-based face recognition to identify registered individuals and record attendance automatically. The proposed workflow includes face detection, image preprocessing, feature extraction, identity matching, duplicate-checking, and storage of timestamped attendance records. OpenCV can support image capture and processing, while a trained recognition model such as FaceNet can generate facial embeddings for comparison. The paper discusses system objectives, relevant literature, a proposed methodology, evaluation measures, privacy safeguards, and future improvements. Since no institutional dataset or live experiment is supplied for this paper, numerical performance results are not claimed; instead, a testing and analysis framework is presented. The system is intended to reduce routine manual work while retaining human review for uncertain matches and attendance corrections.
KEYWORDS: Facial recognition, attendance management, computer vision, machine learning, OpenCV, FaceNet, face detection, biometric security, automated attendance.
A FACIAL RECOGNITION ATTENDANCE SYSTEM
PRERNA CHAVAN


