Development of Driver's Cognitive Ability Training System using 360 Degree Video

2022 Fiscal Year Final Research Report

• Project/Area Number: 19K03086
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 09070:Educational technology-related
• Research Institution: Akita Prefectural University
• Principal Investigator: Saruta Kazuki 秋田県立大学, システム科学技術学部, 教授 (80282193)
• Co-Investigator(Kenkyū-buntansha): 寺田 裕樹 秋田県立大学, システム科学技術学部, 准教授 (40360002)
• Project Period (FY): 2019-04-01 – 2023-03-31
• Keywords: 運転者教育 / 認知能力 / 360度カメラ映像 / 画像認識 / 視線計測 / 認知負荷

Outline of Final Research Achievements

The objective of this research project was to develop a system to improve the cognitive ability of drivers by measuring and analyzing their gazing behavior toward pedestrians and hazardous locations by presenting 360-degree camera images from the driver's viewpoint on a head-mounted display equipped with a gaze measurement function.
The research results include the development of an automatic object detection method that combines object detection technology using deep learning and gaze measurement technology, as well as a system for training the cognitive abilities of drivers. In addition, an environment that collects and presents a variety of video images under different weather conditions and at different times of the day has been constructed, making it possible to analyze the gazing behavior of subjects in the same scene. Furthermore, we clarified differences in gazing tendencies among subjects and under different conditions through gazing behavior analysis experiments.

Free Research Field

情報学

Academic Significance and Societal Importance of the Research Achievements

本研究課題において開発したシステムにより，同一シーンに対して被験者注視行動を分析することができ，かつ360度カメラ映像を用いることでより実走行環境に近く臨場感の高いシーンを提示することも可能とした。また，映像さえ収集すれば様々なシーンを提示できること，注視物体を目視ではなく自動判定する仕組みを導入することもできた。従来の実験環境のように被験者毎に走行シーンにバラつきが生じることや，注視物体を目視で判定することがなく，研究効率を大きく向上する仕組みを実現でき，学術的・社会的な意義の高い成果が得られたといえる。