스테레오비젼기반 승강장 모니터링시스템

초록

Passenger safety is a primary concern of railway system but, it has been urgent issue that dozens of people are killed every year when they are fallen from train platforms. Recently, advancements in IT have enabled applying vision sensors to railway environments, such as CCTV and various camera sensors. In this paper we propose a stereo vision based platform monitoring system which automatically determines in real-time whether human or object is in preset monitoring area or not. The system uses stereo image processing technique employed in ITS technology to improve detection rate. This paper introduces system overview, configuration and detection process of the system. According to the results, we expect the proposed system requirements will play a key role for establishing highly intelligent monitoring system in railway.