<![CDATA[Face recognition and object detection technologies have been used and developed rapidly in various fields such as security, facilities management, and surveillance. Churches, as a place where many people gather, often face challenges in seating management and monitoring congregation attendance, which is still done traditionally or manually. This traditional approach not only requires a lot of time and effort, but is also prone to human error. Therefore, a system was designed to be able to detect the availability of chairs and identify the faces of the congregation automatically, using the YOLOv8 method and a Convolutional Neural Network (CNN) based on the ResNet-50 model for face detection and recognition. The test results from the 3 groups tested obtained an average accuracy of 85.26% and a detection accuracy of 95.46% with the YOLOv8 model training reaching 97% mAP50 and the ResNet50 model with an accuracy of 99.54% and a validation accuracy of 99.37%.]]>
Copyrights © 2025
