Work

SSD-Based Intelligent Invigilation System for Multi-Modal Behavioral Features in Examination Scenarios

Year: 2025

Type: conference-paper

Abstract: Against the background of low efficiency and lack of accuracy of traditional invigilation methods, this paper proposes a solution to design and implement an intelligent invigilation system to improve the level of examination invigilation intelligence. The system includes three user roles: administrators, invigilators, and candidates. Its functional modules include user management, data acquisition, and preprocessing. The core algorithm adopts a Single Shot MultiBox Detector(SSD) target detection algorithm, which is used to analyze and judge the abnormal behaviors of the candidates intelligently. The experimental results show that this intelligent invigilation system solves the problems of traditional invigilation well and effectively improves the intelligence level and management efficiency of exam invigilation.

Source: International Conference on Computer and Multimedia Technology (ICCMT)

DOI: 10.1145/3757749.3757782

Authors: Xinpeng Gao, Shengming Zhang, Run Chen, Weinan Wang, Jiajin Cao, Ningyuan Xu, Ran Tao, Zhihong Pan

Institutions: Guangzhou Xinhua University, Guangdong Polytechnic Normal University, Jilin University

Language: English

Topic: Academic integrity and plagiarism

Subfield: Safety Research

Field: Social Sciences

Domain: Social Sciences

SDG: Peace, justice, and strong institutions

Open Access status: hybrid

OpenAlex ID: W4414182345

FWCI:

Cites:

Cited by:

Related to: