Facial Recognition Attendance Scheme on CCTV Cameras Using Open Computer Vision and Deep Learning: A Case Study of International University of East Africa (IUEA)
Publication Date: 09/08/2022
Author(s): Edison Kagona.
Volume/Issue: Volume 2 , Issue 1 (2022)
Abstract:
There are plenty of home security cameras currently on the market, many allowing you to stream live footage over the internet and receive alerts whenever someone wanders past. This paper discusses the facial recognition attendance scheme on CCTV Cameras using OpenCV and Deep learning. Generally, Face Recognition is a method of identifying or verifying the identity of an individual by using their face. Various algorithms are there for face recognition but their accuracy might vary. In this paper, we discuss how we can do face recognition using deep learning. We used face embeddings to perform deep_metric_learning and the development steps of the scheme were; face detection, feature extraction, and lastly comparing faces. The legacy system used at IUEA Was firstly studied in more detail. More requirements for the proposed system were obtained and the system was developed. The interfaces for the new system were implemented using HTML, Bootstrap, and Django which is a high-level Python web framework. After the implementation, the new system was then tested, validated, and deployed for use.
Keywords:
Facial Recognition, Detection, CCTV Cameras, OpenCV, Deep Learning.
