Year: 2024
Type: article
Abstract: This paper presents the development of a safety check support device aimed at further enhancing platform safety checks using side cameras installed on rolling stock. The device employs deep learning-based image processing techniques to detect passenger approach on platforms in real-time from camera footage, notifying the driver to assist with safety checks. Additionally, the device is equipped with automatic passenger counting functionality at each station. An overview of the developed device and the results of evaluation experiments are provided.
Source: Quarterly Report of RTRI
Authors: Wataru GODA, Hiroki MUKOJIMA, Nozomi NAGAMINE
Institution: Information and Communication Technology Division
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4401798941