Embedded Low Power Blockchain Traceability Solution for University Classroom Attendance

Abstract

The emergence of artificial intelligence and decentralized applications has made it possible to set up efficient systems in terms of traceability. The combined technologies are used in several fields such as fintech, industry 4.0, smart agriculture, etc. Their application can also impact educational and academic fields. There is growing interest in leveraging blockchain technology to securely store and retrieve student records. This paper proposes a new smart system that uses deep learning and blockchain technologies to store and manage student attendance. It relies on an embedded platform based on a camera and a Raspberry PI platform that uses artificial intelligence for face recognition and on the blockchain for secure data storage. Evaluating the resulting model on the Labeled Faces in the Wild (LFW) benchmark yields an impressive accuracy rate of 0.9938 with a standard deviation of 0.00272. Moreover, the proposed system provides a complete and accurate record of the entire student learning process, and, thus, reduces the risk of falsified educational records. It also provides potential benefits to future employers by giving them access to large amounts of verified and systematically accumulated data, allowing them to identify and hire qualified students.