Face Recognition Based Attendance System Using Deep Learning Algorithm

Abstract

The Face Recognition Based Attendance System is an innovative advancement in attendance tracking, combining cutting-edge facial recognition technology with previously established methods. This system provides an effortless, automated, and highly accurate method to attendance tracking, which is especially useful in educational institutions and organizational contexts. This paper includes an in depth look into the use of advanced deep learning models to improve attendance monitoring methods. To construct an extensive facial recognition system, the researchers used a Convolutional Neural Network (CNN), ResNet50, and EfficientNetB7. The system obtains notable accuracy levels on the test set using a thorough technique that includes data collecting, labeling, and model training, indicating its proficiency in recognising persons. The CNN has a high accuracy of 98.61, demonstrating its powerful facial recognition skills. While ResNet50 and EfficientNetB7, although having lesser accuracies of 88.09% and 63.43%, respectively, provide useful insights into the relative performance of alternative deep learning architectures. The research goes beyond technology to explore ethical concerns, societal impact, and sustainability over the years.