Facial recognition, a branch of image processing, is widely used in attendance systems to improve efficiency and security. This study develops an employee attendance monitoring system that integrates facial recognition using the Eigenface algorithm in OpenCV. The system records each individual's facial data alongside a password, enabling automated attendance tracking. Testing results indicate that with a database of 10 facial entries, the system achieved 100% accuracy in recognizing individuals. However, as the database expanded beyond 10 entries, accuracy declined to 80%, influenced by factors such as lighting variations, differences in facial structures, and increased data volume. This study employed a Research and Development (R&D) methodology, with expert validation yielding a score of 3.4, categorizing the system as "Highly Valid." User testing with 11 participants resulted in an overall score of 36, classifying the system as "Very Good (Valid)." The findings highlight the potential of facial recognition in improving attendance management while minimizing fraudulent entries. Future research should focus on optimizing recognition accuracy in larger databases through refined preprocessing techniques, image quality adjustments, and deep learning models.
