Author(s)

Hengyi Guo, Xueyan Cao

Corresponding Author

Hengyi Guo

ABSTRACT

As an important branch of artificial intelligence, face recognition (FR) technology has the characteristics of anti-counterfeiting and non-repudiation. Applying this technology to libraries can meet the needs of libraries to fast and accurately identify readers. In this work, we investigate the FR system deployed in the library of the University Electronic Science and Technology of China (UESTC), from the perspective of data mining. After properly modeling the operating data, we propose certain parameters for evaluating the operating status of the FR system, and illustrate how to adjust the related threshold to maintain a stable operation of the FR system. The analysis method presented in this work can be used as a reference by other libraries when deploying FR systems.

KEYWORDS

Access control system, face recognition, library, statistical analysis