Faculty of Computing, Library and Information Science (FCLIS)

Permanent URI for this collection

http://hdl.handle.net/20.500.12493/222

Browse

Browsing Faculty of Computing, Library and Information Science (FCLIS) by Subject "Face Recognition"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    ItemRestricted
    Enhancing Attendance Management Through Face Recognition Technology: A Case Study at Rugarama School of Nursing and Midwifery.
    (Kabale University, 2024) Taremwa, Benjamin
    Inaccuracies and inefficiencies in traditional attendance management systems have long posed challenges for educational institutions. This study, titled "Enhancing Attendance Management through Face Recognition Technology: A Case Study at Rugarama School of Nursing and Midwifery," aimed to develop a more accurate and efficient solution using Local Binary Pattern Histogram and Convolutional Neural Networks algorithms to automate attendance tracking.A mixed-method approach was employed, combining system testing with user feedback from administrators, staff, and students. The system was evaluated on key metrics such as accuracy and time efficiency, achieving an 80% accuracy rate and full agreement on time efficiency (100%). The study also highlighted challenges related to lighting conditions and privacy concerns, which impacted system performance in real-world conditions. The system’s integration with the school’s course schedules provided reliable, real-time attendance tracking and significantly reduced errors and manipulation associated with traditional methods.This research contributes to filling a notable gap in the literature regarding the use of face recognition systems for course-specific attendance determination in educational settings. Despite the system’s effectiveness, future research should explore improvements in image quality across diverse environments and investigate other biometric methods to enhance security.The findings of this study suggest that face recognition-based attendance systems have the potential to revolutionize attendance management not only in education but also in other sectors where identity verification is critical, such as healthcare and corporate environments. However, limitations such as lighting variability and dataset size indicate further refinements are needed to optimize the system for broader implementation.