Design and Development of Integrated Human Resource Management System with Face Recognition Attendance

Aim:

            Our study aims to design and develop a smart human resource and attendance management system using facial recognition technology to automate employee identification and attendance tracking. The system leverages AI-based face detection to ensure accuracy, eliminate proxy attendance, and streamline HR processes, thereby improving workforce efficiency, data security, and transparency within the organization.

Abstract:

            The Human Resource and Attendance Management System Using Facial Recognition aims to automate employee attendance tracking with improved accuracy, security, and efficiency. Traditional methods like manual entry or RFID cards often suffer from inaccuracies and proxy attendance. To overcome these challenges, the proposed system integrates advanced deep learning models — ResNet-10 SSD (deploy.prototxt and res10_300x300_ssd) for face detection and SFace (face_recognition_sface.onnx) for face recognition. These models work together to detect and identify faces with high precision, even in complex environments. Unlike earlier approaches such as YuNet, SFace (standalone), and KNN classifiers, the proposed combination provides superior performance by accurately handling different facial expressions, low-light conditions, and varied face angles. The system ensures real-time recognition with minimal errors, automatically marking attendance once a valid face is detected. Additionally, it supports essential HR features like Leave and Salary Management and an Attendance Report Dashboard for date-wise monitoring. By integrating robust AI-based face recognition with HR automation, this system enhances transparency, eliminates manual effort, and delivers a reliable and contactless attendance solution suitable for modern organizations and institutions.

Proposed System:

            The proposed system enhances accuracy and robustness by integrating deep learning–based models — ResNet-10 SSD (deploy.prototxt and res10_300x300_ssd) for precise face detection and SFace (face_recognition_sface_2021dec) for advanced face recognition. The detection model identifies and localizes faces in various lighting and pose conditions, while the recognition model extracts deep facial features for reliable identity verification. Together, these models overcome the limitations of earlier methods, offering superior performance in low-light environments, different face angles, and varied facial expressions. The system automatically marks attendance once a valid face is recognized and securely stores data for HR processing. It also supports additional modules such as Leave and Salary Management and an Attendance Report Dashboard for real-time analytics and monitoring.

Advantages:

            The proposed system delivers high accuracy and adaptability by leveraging deep learning–based facial recognition. It can efficiently handle complex conditions like lighting variations and side-face detection, ensuring consistent results. The models used in the system provide faster detection, improved recognition precision, and better scalability compared to traditional approaches. Furthermore, it eliminates proxy attendance, reduces manual effort, and enhances transparency in attendance records. By combining AI-powered recognition with HR automation, the system ensures a secure, contactless, and efficient solution for modern workforce management.