Development of Information Infrastructure Technology to Enhance the Safety and Quality of Childcare Workers

2023 Fiscal Year Final Research Report

• Project/Area Number: 19K02647
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 09030:Childhood and nursery/pre-school education-related
• Research Institution: Surugadai University (2023); Tokyo University of Social Welfare (2019-2022)
• Principal Investigator: Tachi Hidenori 駿河台大学, メディア情報学部, 講師 (90402148)
• Project Period (FY): 2019-04-01 – 2024-03-31
• Keywords: 安全管理 / 保育 / アノテーション / 気づき / 保育士 / 暗黙知

Outline of Final Research Achievements

.In this study, we aimed to develop a video analysis system specifically focused on the behavior of young children. Our objectives included tagging children's movements in a daycare setting, distinguishing between caregivers and children, and developing a scene detection system. We also improved and evaluated an annotation system that allows users to highlight noteworthy points during video playback. Additionally, we collaborated with daycare staff and educators in training programs to classify situations and levels where caregiver intervention or verbal interaction was necessary. Furthermore, to digitize 42 years of conference proceedings from the Japan Society of Child Science, we converted paper-based documents into digital text and extracted terms related to risks and near-miss incidents in daycare settings. We analyzed the frequency and relationships of these terms with other words.

Free Research Field

教育工学

Academic Significance and Societal Importance of the Research Achievements

保育経験者が暗黙知をもとに行ってきた潜在的な知見や判断を分析し，システムに取り入れ，日々の活動において，幼児の危険を部屋のカメラからリアルタイムに察知し保育士に知らせ事故を未然に防ぐサポートを行う。また，撮影後の定点カメラ等から研修・教育用の動画教材としての場面の抽出を容易にすることで，教材不足を解決する。ただし，あくまでもシステムは保育士をサポートするための補助ツールであり，最終的に判断するのは保育士自身である。また，子どもの動態に特化したデータセットを作成することで，様々な研究やシステムへの応用が可能となる。