17:15 〜 18:45
[G03-P05] A Pilot Study of the Generation of Local Signs or Educational Materials for Disaster Prevenstion using AI Image Generator
キーワード：AI、防災教育、情報デザイン
The Noto Peninsula Earthquake of 2024, which occurred on January 1, 2024, registered 7 on the Japanese seismic intensity scale, and as of January 31, 2024, 238 people had been killed. In this earthquake, due to delays in assessing the damage and errors in judgment by the national and prefectural governments, the establishment of an initial response system, the importance of which has been recognized since the Great Hanshin-Awaji Earthquake, did not proceed smoothly, resulting in delays in rescue operations and a lack of support for disaster victims due to the shutdown of volunteers. On the other hand, in a district where evacuation drills have been conducted for more than 10 years, the death toll from the tsunami that struck immediately after the earthquake was successfully reduced to zero, and the importance of correct evacuation behavior has once again become widely known to the public. In order to master smooth evacuation behavior, training and education need to be routinely provided throughout this country. However, priorities for disaster prevention behavior and evacuation routes vary depending on local characteristics such as residential topography and building structures, and uniform signs and displays do not provide sufficient training and education. On the other hand, disaster prevention specialists and teachers need to acquire a variety of skills to create original signs and posters that take regional backgrounds into account. This study focuses on image generation by AI, which has been rapidly developed in recent years. For the purpose of promoting of disaster prevention educations, the possibility of AI creating disaster prevention posters and evacuation signs that take local circumstances into account, in place of disaster prevention staff or teachers, is discussed. This work was supported by JSPS KAKENHI Grant Number JP23K02821.