FaceAttendance: Leveraging Facial Features for Attendance Tracking System
Suja A. Alex, Ângela Maria Alves, Gabriel Gomes de Oliveira, Gabriel Caumo Vaz, Eric CK Cheng · 2024
Biometric authentication verifies a person's identity using distinct biological characteristics such as fingerprints, iris patterns, facial features, or behavioral traits such as typing rate. It presents a highly secure and convenient means of access control and identity confirmation, given the difficulty in replicating or counterfeiting these traits. During authentication attempts, a biometric system captures the individual's biometric data, compares it with stored templates, and grants access upon a successful match. This technology sees widespread adoption across various sectors including smartphones, border control, banking, and healthcare. A secure and simplified method of student attendance tracking is still challenging. This work introduces an innovative approach for automated classroom attendance tracking by precisely identifying students' facial features using a high-accuracy facial recognition system. This enables the proposed system by leveraging deep learning algorithms for facial feature extraction and recognition, ensuring seamless and contactless attendance recording. This simplifies attendance tracking in a classroom by precisely identifying students' facial features and recording their attendance. The evaluation used a dataset of 10,000 facial images representing 100 unique students. The dataset includes images captured under different lighting conditions, angles, and varying facial expressions to simulate a realistic classroom environment. The proposed work achieved an overall accuracy of 99.01%, indicating that 99.01% of the students were correctly identified and their attendance was accurately recorded.