This research presents an intelligent approach for authenticating student IDs. To ensure that students are who they say they are, it employs deep learning and facial mapping. You may now use your face instead of a physical ID card using this technique. Both the process\'s security and its convenience are enhanced by this. Access is granted via ID cards in conventional systems. These may be misplaced, taken, or abused. Using state-of-the-art technology, our solution resolves these issues. We use student face data to construct a deep learning model. At the entrance, a camera records the face of every student. Using a database comparison, the system verifies the current picture. The gate will open if the faces match. If it doesn\'t, the security team will get a notification. Regardless of the ambient light or the user\'s emotional state, the device performs well. Additionally, it is quick, precise, and doesn\'t need any physical touch. That means ID cards won\'t be needed as often. Additionally, it lessens the possibility of unlawful entry.
Introduction
The text discusses the integration of biology, statistics, and computer science in facial recognition technology, highlighting its evolution from simple statistical models to sophisticated deep learning systems. Facial recognition involves two key stages: face detection and face recognition, with modern systems achieving up to 99.15% accuracy. Unlike other biometric methods, it is contactless, user-friendly, and can operate remotely, making it ideal for applications like surveillance and authentication.
Literature Review:
Various systems use techniques like deep neural networks, SVM classifiers, PCA, and LDA for face recognition.
Systems tested on small datasets show varying accuracies (around 60-83%).
Limitations include sensitivity to lighting, location changes, and scalability issues.
Existing Systems:
Use conventional cameras that struggle in poor lighting.
Lack alerts for unauthorized individuals.
Proposed System:
Uses cameras or sensors for input (images or video).
Employs Python-based classifiers, OpenCV, MTCNN for face detection, and FaceNet for recognition via 128-dimensional embeddings.
Incorporates data augmentation to enhance training with a dataset of 30 students expanded via image transformations.
Stores attendance and alerts on unrecognized faces via buzzer alarms and emails.
Uses QR codes for visitor access, enhancing security and access control.
Includes real-time notifications to coordinators if students leave during restricted times.
Modules:
Face data collection with augmentation.
Real-time face recognition.
Automated gate authentication.
Access control with QR codes.
Timestamp logging of arrivals/departures.
Alerts for security breaches.
System Architecture & Results:
The system features a GUI interface, registration module, database storage, face recognition with voice feedback, CSV reporting, and QR code generation for visitors.
Conclusion
A potential solution to the problems of accessibility and security in student housing might be an admission system that relies on facial recognition technology. They may assist keep dorms safe from intruders and expedite the entry and leave procedures. The ultimate goal of this study is to design a smart hall security system that can recognize faces in real-time by using intelligent face detection and recognition technology. A real-time face recognition system with detection and recognition capabilities, suitable to multiple security services, was created after this research studied numerous facial recognition algorithms. The system is simple to set up, affordable, and straightforward for users. Updating student datasets to make sure they are secure and accurate could be a part of future improvements. It is possible for security staff to get push alerts whenever an unfamiliar individual is identified. Furthermore, novel methods are needed to circumvent constraints and successfully identify many people in diverse settings.
References
[1] Khairuddin, MH, and Shahbudin SK, Kassim M. \"A Smart Building Security System with Intelligent Face Detection and Recognition.\" In IOP Conference Series: Materials Science and Engineering, 2021;1176(1):012030. IOP Publishing.
[2] Chauhan, M, and Sakle M. \"Study & Analysis of Different Face Detection Methods.\" International Journal of Computer Science and Information Technologies, 2014;5(2):1615- 1618.
[3] Aydin, Ilhan, and Nashwan Adnan Othman. \"A New IoT Combined Face Detection System UsingComputer Vision for Security Application.\" In 2017 International ArtificialIntelligence and Data Processing Symposium (IDAP), 2017;1-6. IEEE.
[4] Raghuwanshi, Anshun, and Preeti D. Swami. \"An Automated Classroom Attendance System Using Video-Based Face Recognition.\" In 2017 2nd IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), 2017;719-724. IEEE.
[5] 28. Soniya V, Swetha Sri R, Swetha Titty K, Ramakrishnan R, Sivakumar S. \"Attendance Automation Using Face Recognition Biometric Authentication.\" In 2017 International Conference on Power and Embedded Drive Control (ICPEDC), 2017:122-127. IEEE.
[6] Yang, M. H., Ahuja, N., & Kriegman, D. (2000). Face detection using mixtures of linear subspaces. In Automatic Face and Gesture Recognition, 2000. Proceedings. Fourth IEEE International Conference on (pp. 70-76). IEEE.
[7] Yang, G., & Huang, T. S. (1994). Human face detection in a complex background. Pattern recognition, 27(1), 53-63.
[8] N.Stekas, D.Heuvel, “Face Recognition Using Local BinaryPatterns Histograms (LBPH) on an FPGA-Based System on Chip (SoC),” IEEE, Chicago, IL, USA, Aug 2016.