Identity Verification Using Face Recognition for Artificial-Intelligence Electronic Forms with Speech Interaction

摘要

Concern over the decline in Japan's manufacturing competitiveness has increased in recent years. In particular, falsification of inspection data is a social problem that could undermine Japan's manufacturing industry, which is founded on a dedication to high quality. Falsification could be prevented by ensuring transparency of the inspection process by visualizing the process. End-to-end visualization facilitates early detection and prevention of various law infractions. In the workplace, visualization requires an efficient low-cost identity-verification method that ensures ease of visual confirmability for product traceability. We previously developed Al-forms, i.e., artificial-intelligence electronic forms, that provides a speech interface as a means of improving the standard work process in workplaces by making operations more efficient and visualizing processes. Al-forms improves production efficiency and visualizes the collected operation records by enhancing the readability and writability of records and handover operations that are not sufficiently supported by traditional electronic forms. To prevent falsification of inspections, it is necessary to use a widely deployed device and verification method in the workplace. We propose an identity-verification method for applying face recognition to Al-forms and developed a smartphone app for Al-forms. Preliminary feasibility testing involving 11 workers in an actual workplace confirmed that identity verification is possible when face recognition is carried out with frontal images of workers who are not wearing face masks. The face-recognition process completed within 0.4 s, enabling workers to seamlessly begin work with Al-forms. Recording both collation photos and worker names during identity verification also made it possible for a human to visually confirm a worker's identity. Discussion with workers and supervisors after the feasibility tests provided findings for improving our face-recognition app for closer integration of Al-forms and our identity-verification method at arbitrary times.