Abstract—This article explores methods for improving the quality of existing facial biometric recognition systems by fine-tuning the model on new data. It examines the overall framework reflecting the fundamental operating principle of the biometric identification security system, as well as the main approaches and methods for addressing this task using the Deep Neural Network (DNN) face detection method in OpenCV. A facial recognition software suite has been developed, which includes: a detection module, a head position determination module, a user identification module, an Access Control and Management System (ACMS) module, and a training module. Research on existing methods to enhance the accuracy of identification algorithms and systems has been conducted. An analysis of the increase in recognition rates after system fine-tuning for different times of day was performed. The results of the study showed that the developed module ensures high accuracy and reliability. The recognition rate increased by approximately 4–5% as a result of system fine-tuning. Additionally, it is worth noting that ACMS with facial recognition technology represents a powerful tool for educational institutions seeking to automate their attendance tracking processes. This step marks significant progress in applying advanced technologies to increase the efficiency and accuracy of attendance management.
 
Keywords—identification, recognition systems, recognition algorithms, Deep Neural Network (DNN) face detection method in OpenCV, fine-tuning, integration with Access Control and Management System (ACMS)


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).