期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：1
页码：1204
DOI：10.15680/IJIRCCE.2017.0501147
出版社：S&S Publications
摘要：Railway is major part of India and it is being the cheapest way of transportation today. Daily newspapersshow different accidents in the railway track. Railway accidents caused by obstacles are the most important issues thatshould be solved. There are many methods used to detect obstacles in the railway track. This paper gives an idea aboutvarious methods of obstacle detection in railway track. Majority of methods use IR sensors to detect the obstacle.Machine learning approach does not detect obstacles with unknown classes. So it uses detecting obstacles bycomparing input and reference train frontal view camera images. In Indonesia method used is by using infrared systemwhich is placed in front of the locomotive, so when there are objects detected within a range distance, the indicatorlights will light up. So that each method has its own advantage and disadvantages.
关键词：IR sensors; Obstacle; Machine learning; Camera images; infrared.
