Smart Attendance Management System Using Data Augmentation

Abstract

Due to the enhanced representational capabilities, Convolutional Neural Networks (CNN) have evolved to be a potent method for image processing and have made considerable improvements in face recognition lately. On the other hand, deep learning techniques frequently struggle to generalize small and sparse dataset samples. Thus generalizing new categories of data using data augmentation is a workable leveraging strategy to deal with the issue of few training samples. This technology can assist in resolving the issues with the current attendance monitoring process. In order to overcome the issues with conventional attendance monitoring systems in educational institutions, this research introduces a revolutionary smart attendance management system that combines data augmentation and convolutional neural networks (CNN) to improve the accuracy and resilience of the system, utilizing deep learning algorithm to recognize and validate faces collected through a camera. The suggested system provides an easy-to-use platform to monitor attendance records and is extremely scalable, adaptable, and user-friendly. The experimental findings of this model illustrate the better performance of the proposed system, exhibiting its capacity to precisely identify and verify student attendance and exceeding the performance of the traditional attendance management systems. Overall, this system reduces manual effort by the administration and boosts student engagement to improve attendance monitoring procedures in educational institutions.