Objective: Recording attendance is a critical process in academic institutions due to its significant impact on student performance and engagement. Current methods for recording attendance are often time-consuming and labour-intensive for lecturers and administrative staff, necessitating the development of more efficient and flexible solutions. While various automated attendance systems have been proposed, they often encounter challenges related to cost, implementation complexity, or reliability, hindering widespread adoption in educational settings. Method: This paper introduces a novel approach to automating attendance registration using face recognition technology. Our method integrates multiple feature extraction algorithms within a user-friendly graphical interface, specifically designed in English to enhance usability. By using existing security cameras commonly found in academic institutions, our approach addresses both cost and time inefficiencies. The attendance registration process involves capturing a video of the classroom, which is then processed to identify and log student attendance in a CSV file. A significant aspect of our study is using a comprehensive dataset comprising 2,170 images collected from 31 students at Mustansiriyah University. This extensive dataset enhances the robustness and reliability of our system, providing a diverse range of facial expressions, angles, and lighting conditions that improve the accuracy and generalizability of our model. Result: The system demonstrated accuracy of up to 100%, with deep learning algorithms outperforming machine learning methods. Conclusion: These promising results suggest that face recognition technology can effectively streamline and automate attendance tracking, offering a viable solution for educational institutions seeking to improve operational efficiency and accuracy

Objective: Recording attendance is a critical process in academic institutions due to its significant impact on student performance and engagement. Current methods for recording attendance are often time-consuming and labour-intensive for lecturers and administrative staff, necessitating the development of more efficient and flexible solutions. While various automated attendance systems have been proposed, they often encounter challenges related to cost, implementation complexity, or reliability, hindering widespread adoption in educational settings. Method: This paper introduces a novel approach to automating attendance registration using face recognition technology. Our method integrates multiple feature extraction algorithms within a user-friendly graphical interface, specifically designed in English to enhance usability. By using existing security cameras commonly found in academic institutions, our approach addresses both cost and time inefficiencies. The attendance registration process involves capturing a video of the classroom, which is then processed to identify and log student attendance in a CSV file. A significant aspect of our study is using a comprehensive dataset comprising 2,170 images collected from 31 students at Mustansiriyah University. This extensive dataset enhances the robustness and reliability of our system, providing a diverse range of facial expressions, angles, and lighting conditions that improve the accuracy and generalizability of our model. Result: The system demonstrated accuracy of up to 100%, with deep learning algorithms outperforming machine learning methods. Conclusion: These promising results suggest that face recognition technology can effectively streamline and automate attendance tracking, offering a viable solution for educational institutions seeking to improve operational efficiency and accuracy