A School Safety System for Detecting Threatening Individuals and Providing Escape Routes via CiRA CORE Platform

Abstract

This research aims to develop and evaluate the effectiveness of a system for detecting dangerous individuals and recommending escape routes within a school environment using the CiRA CORE platform. The system processes images from CCTV cameras with the YOLO V4-tiny model to detect weapons and employs Dijkstra's algorithm to calculate the most suitable escape route. The results are displayed through a real-time notification system. The research findings revealed that: Part 1 The YOLO V4-tiny model has high performance in terms of speed, resource efficiency, and accuracy in detection. Part 2 Dijkstra's algorithm is efficient for calculating the shortest distance for escape routes. Part 3 The collaboration between the CiRA CORE platform and Line Notify can accurately notify the location and escape route with 100 percent precision. Part 4 The collaboration between the CiRA CORE platform and the web application can accurately identify the location and escape route with 100 percent precision.