Traditional classroom monitoring requires manual supervision, leading to inefficiency, proxy attendance, and heightened faculty workload. This paper proposes an intelligent, AIenabled smart classroom system integrating Edge AI, embedded systems, and IoT to automate attendance, enhance security, and optimize energy management. The proposed architecture employs K-Nearest Neighbor (KNN) based live face recognition to accurately record student attendance, entirely eliminating manual errors and proxy entries. Additionally, the system incorporates RFID door authentication for secure access, alongside Laser and Light Dependent Resistor (LDR) sensor arrays to detect and alert regarding unauthorized student movements. To address power consumption, Passive Infrared (PIR) sensors dynamically manage classroom lighting based on real-time occupancy. Powered by Flask for web integration, OpenCV for image processing, and managed via Arduino and PIC microcontrollers, the system outperforms legacy 8051-based surveillance architectures by providing real-time data logging, higher accuracy, and automated energy-saving mechanisms. Experimental results demonstrate a 100% operational success rate in integrating hardware and software environments, providing a highly reliable, cost- effective, and scalable solution for modern educational institutions.

The proposed project successfully integrates AI and embedded systems to create an autonomous, smart classroom automation platform. It achieves highly accurate face-based attendance logging, eliminates the potential for proxy entries, and guarantees secure physical access through multi-factor (Face + RFID) capabilities. Real-time perimeter monitoring enforces classroom discipline, while the PIR-driven electrical layout ensures sustainable energy utilization.

[1]Smith, J., “Impact of Computer Vision on Educational Record Keeping,” Journal of Smart Environments, vol. 12, pp. 45-56, 2024. [2]Lee, K., and Patel, R., “Analysis of Optical Tripwires in Surveillance Contexts,” Sensor Networks IEEE, 2023. [3]Nguyen, T., “Energy Optimization in Smart Buildings Using Passive Infrared Topologies,” Automation in Construction, 2022 [4]Kim, H., “Exploration of RFID Access Control in Institutional Environ ments,” Journal of Embedded Security, 2021. [5]Deepika R., Shalini P., et al., “Edge AI Development of Sustainable IoT Frameworks,” IEEE Access, 2024. [6]Jinpeng Miao et al., “A Microservice-Based Smart System to Detect Intrusions at the Edge,” IoT Systems Journal, 2024. [7]Ashish Gawande, “Smart Intrusion Detection Using Multi-Modal IoT and Edge AI,” 2025. [8]Miroslaw Hajder et al., “AI-Based Integrated Multi- Sensor System with Edge Computing,” Tech. Review, 2025 . [9] Tsheten Dorji et al., “AI-Based Intrusion Detection System,” Intelligent Systems, 2025. [10]Subha N. et al., “AI-Driven Vision Systems for Sustainable Monitoring,” IEEE Trans. on AI, 2025. [11]Kong Ka Hing and Mehran Behjati, “Edge Intelligence for Conservation Using TinyML,” 2025. [12]Golam Sarowar et al., “Enhanced Wireless Control System for Hardware Modules,” Embedded Networks Vol. 5, 2013

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close