REAL TIME AUTOMATED ATTENDANCE SYSTEM USING FACIAL RECOGNITION AND MACHINE LEARNING

Abstract

In order to replace manual approaches, this study proposes an automated biometric attendance management system that uses facial recognition technology. Traditional attendance methods consume significant time, are prone to human errors, and may interrupt classroom activities. The proposed system utilizes a fixed surveillance camera to capture real-time images of students and applies image processing techniques such as Haar Cascade for face detection and LBPH (Local Binary Pattern Histogram) for face recognition. The extracted facial features are matched with a pre-stored database to accurately identify individuals and automatically record attendance. The system operates without human intervention, ensuring a contactless and efficient process. It significantly reduces the time required for attendance marking while improving overall accuracy and reliability. Unlike traditional methods, the proposed system can detect and recognize multiple faces simultaneously, allowing attendance to be marked for more than one person at a time. The system compares detected faces with a pre-trained dataset and marks attendance automatically when a match is found within a predefined confidence threshold. Additionally, the system can generate attendance logs and maintain structured records. The proposed solution minimizes proxy attendance, enhances classroom management, and provides a scalable approach for educational institutions.