Karki, Nischal, Anwar, Aamir, Rehman, Ikram ORCID: https://orcid.org/0000-0003-0115-9024, Husamaldin, Laden and Saadati, Parisa (2023) Smart Attendance Monitoring System Using Face Recognition for People with Disabilities (PwDs). In: 2023 IEEE International Smart Cities Conference (ISC2), 24-27 Sep 2023, Bucharest, Romania.

Full text not available from this repository.

Abstract

This research presents an innovative and inclusive solution to automate attendance-taking processes in various settings, particularly educational institutions, by introducing a user-friendly GUI-based smart attendance monitoring system utilising facial recognition technology. Addressing the limitations and inefficiencies of traditional methods like RFID cards or manual signing, the proposed system offers an efficient, accurate, and accessible means of registering individuals' attendance, specifically benefiting students with disabilities. Leveraging advanced models like the pre-trained DNN Caffe model for face detection and the Openface model for face recognition, complemented by an SVM classifier and Tkinter for the User Interface (UI), the system achieves an impressive average accuracy of 92% in diverse contextual parameters, including challenging lighting, occlusion, orientation, and background conditions. Notably, the model can recognise side-oriented faces and multiple faces simultaneously, making it robust for real-world environments. This research project contributes to attendance management practices, highlighting the benefits of the developed system while offering potential solutions to overcome its limitations. It serves as a foundation for further research and opens up possibilities for inclusive attendance tracking in diverse contexts beyond education.

Item Type:	Conference or Workshop Item (Paper)
ISSN:	2687-8860
ISBN:	9798350397758
Identifier:	10.1109/ISC257844.2023.10293664
Identifier:	10.1109/ISC257844.2023.10293664
Subjects:	Computing > Information security
Depositing User:	Marc Forster
Date Deposited:	09 Dec 2024 09:40
Last Modified:	09 Dec 2024 09:40
URI:	https://repository.uwl.ac.uk/id/eprint/12976

Actions (login required)

View Item