Design and Research on the Use of AI-Based Image Recognition Technology for Collaborative Campus Safety Management

Abstract

Several significant issues that campuses have long faced include school bullying and unauthorized individuals intruding on school premises. Similar incidents frequently appear in the news, and bullying events can cause severe physical and psychological harm to both students and teachers. The unauthorized entry of unknown persons into schools poses potential dangers to everyone on campus, as no one can predict their intentions. Despite schools' efforts in promoting anti-bullying measures and safety awareness, the effectiveness of preventing such incidents remains limited without concrete actions or measures. Therefore, I believe that substantive protective measures are necessary to further strengthen campus safety and create a secure learning environment.

This study explores the integration of the Internet of Behaviors (IoB) and AI technologies for behavioral recognition within the campus. The research focuses on two key aspects. First, we collect real-time image data from sensors, cameras, or surveillance systems to address school bullying. Using IoT for data transmission, this information is analyzed, and image recognition combined with deep learning techniques is employed to determine if physical bullying is occurring. Once an incident is detected, an immediate alert is sent to the school administration for prompt action, thereby preventing bullying in real time. Second, the prevention of unauthorized individuals entering the campus is achieved through similar IoT and image recognition methods. While ensuring campus safety, it is also crucial to consider the privacy of staff and students and ensure that the application of these technologies remains within legal boundaries.