Edge AI-Powered Access Control Systems for Smart Classrooms
摘要
With technology advancing rapidly, numerous systems aimed at constructing and enhancing Smart IoT Systems are under development. Concurrently, addressing the need for efficient check-in/check-out procedures within organizational spaces globally has sparked significant interest in Smart Access Control solutions. Artificial Intelligence (AI), a burgeoning and disruptive technology, is increasingly integrated into implementing Smart Access Control systems. One approach involves deploying three models-face detection, face recognition, and face anti-spoofing-leveraging the computational power of Jetson Xavier NX for accurate person detection. This paper presents the design and execution of a comprehensive edge/AI system, encompassing software and hardware components, to establish a fully functional Smart Access Control system. Complementary mobile and web applications cater to standard users and organizational owners/administrators. Harnessing edge computing and AI, our system identifies individuals accessing organizational spaces in real-time, seamlessly updating events on the backend server. Experimental findings demonstrate superior processing throughput and energy efficiency of our Jetson Xavier Edge computing platform compared to the Raspberry Pi 4 edge platform.