International Journal of Computer Techniques Volume 12 Issue 4 | Automated Attendance Tracking via Multi-Face Recognition and Intelligent Detection
Automated Attendance Tracking via Multi-Face Recognition and Intelligent Detection
Authors:
Rashmi Mishra R, MCA Student, School of Science and Computer Studies, CMR University, Bangalore, India (Rashmi.mishara@cmr.edu.in)
Dr. Umadevi R, Associate Professor, School of Science and Computer Studies, CMR University, Bangalore, India (Umadevi.r@cmr.edu.in)
Abstract
This study introduces a robust attendance system integrating HOG and Haar-based face detection pipelines with FaceNet-powered deep learning recognition. Real-time performance and multi-face robustness are achieved using preprocessing enhancements like CLAHE and Gaussian filtering. The system processes over 70 individuals per frame with 25ms speed, registering attendance in a structured database and issuing automated PDF reports via email.
Keywords
Multi-face recognition, FaceNet, HOG, Haar cascade, Deep Learning, CLAHE, Attendance Automation, CNN, Real-time Detection
Conclusion
The system effectively automates attendance through hybrid recognition models, maintaining high accuracy under occlusion and lighting variations. Its modular scalability and intuitive dashboard improve management efficiency. Compared to traditional systems, this solution enhances precision and reliability in educational and enterprise contexts.
References
- A. Rao, “AttenFace: A Real Time Attendance System using Face Recognition,” arXiv:2211.07582, 2022.
- A. Touzene et al., “An Embedded Intelligent System for Attendance Monitoring,” arXiv:2406.13694, 2024.
- A. P. Chawla et al., “Multi-face Recognition System for Real-Time Attendance,” ICDLAIR, 2025.
- H. Amri, “CNN Mobilenet and MTCNN Attendance System,” JTOS, 2025.
- R. Krishna et al., “Multiple Face Recognition Attendance System Using Deep Learning,” IJERT, 2023.
Post Comment