A new approach for smart attendance system based on improved video facial recognition technology for smart university

El-Mashad, Yehia and Ali, Hesham A. (2024) A new approach for smart attendance system based on improved video facial recognition technology for smart university. In: Agria Média 2023 : „A magas szintű digitális kompetencia a jövő oktatásának kulcsa”. Eszterházy Károly Katolikus Egyetem Líceum Kiadó, pp. 75-95.

Preview

Text 77_El Mashad.pdf - Published Version Download (1MB) | Preview

Abstract

Since entering the information age, there have been considerable developments in the methods of managing the various learning processes, so there is no need to rely on a large amount of human resources to collect and analyze data. Many technologies have emerged that are capable of analyzing different types of data and providing interdependence and movement to data effectively. So, it can be said that digital transformation has played a decisive role in developing management systems in smart universities. Attendance systems through facial recognition may be considered the most important operation in the smart university. The main objective of this paper is to introduce the attendance system through a new methodology for detecting and identifying faces through video cameras based on artificial intelligence techniques to predict the face and match it with what is in the database. By developing a robust attendance system using video facial recognition technology, the proposed methodology in this paper aims to improve the accuracy, efficiency, and safety of attendance tracking in smart universities. To achieve the proposed goal, this paper will focus on developing a facial recognition algorithm that can accurately identify individuals under varying lighting conditions and facial expressions. The proposed system can provide real-time attendance information, allowing for timely interventions and support for students who may need it. Moreover, the use of video facial recognition technology can help reduce the workload for teachers and administrators. The proposed algorithm is tested, and the experimental results prove that, due to minimal error, better classification accuracy and high confidence value are achieved.

Actions (login required)

Edit Item