Facial Capturing Attendance Tracking System Using Computer Vision

Traditional attendance systems, such as manual roll calls and biometric methods, have drawbacks such as being time-consuming, prone to human error, and vulnerable to proxy attendance. Additionally, faculty members manually updating attendance in Excel increases workload and the risk of mistakes. The proposed facial recognition-based attendance system eliminates manual attendance entry and enables easy web-based access, providing an efficient solution for attendance management. Using webcams, the system ensures a contactless and effective process, reducing manual data entry through automatic attendance reports exportable in multiple formats. This lightens the faculty’s workload while enhancing accuracy. The system employs HaarCascade for face detection and LBPH for face recognition to ensure high accuracy. It also provides real-time tracking, secure data storage, and web-based accessibility, making it scalable across multiple classrooms and departments. To enhance performance and scalability, the system is deployed on AWS cloud infrastructure, leveraging Amazon EC2 for hosting the web application, Amazon RDS for managing the MySQL database, and Amazon S3 for storing captured images securely, excel sheets and yml documents. By integrating these AWS services, the system ensures reliability, security, and seamless accessibility, making it a robust solution for modern attendance management. Key Words: Facial recognition, LBPH, Haar Cascade, Attendance System, Computer Vision, Face Detection, AWS