Fan, Kaiyuan
Article History
Received: 9 May 2025
Accepted: 17 November 2025
First Online: 29 November 2025
Competing interests
: The authors declare no competing interests.
: This study utilized previously collected data from Jiaozuo Normal College, Henan Province, China. The data, consisting of facial images and related metadata, was collected between January 2022 and March 2023 as part of an internal research initiative led by the Department of Information Engineering. As the corresponding author and a faculty member, we confirm that we have the legal and institutional authority to use this data for research purposes, as granted by the Jiaozuo Normal College Research Policy. Informed consent was obtained from all participants, with explicit explanations of data usage, potential risks, and withdrawal rights. Institutional approval was secured from the Jiaozuo Normal College Academic Member Association college’s ethics committee. All methods complied with institutional, national (e.g., China’s Personal Information Protection Law), and international guidelines (e.g., GDPR principles for data protection). To ensure privacy, data was anonymized by removing identifiable information, stored on secure servers with access limited to authorized researchers, and encrypted during transmission. Potential biases were mitigated through dataset diversity checks (e.g., balanced representation across age, gender, and ethnicity) and bias audits during model training to prevent discriminatory outcomes. For ethical deployment in educational settings, the system emphasizes transparency (e.g., notifying users of monitoring), equity (e.g., avoiding disproportionate impact on minority groups), and minimal intrusiveness, aligning with principles to foster trust and prevent misuse as surveillance tools.