DocumentCode :
3120668
Title :
Development of Event Data Recorder for Pedestrians for analysis of traffic accidents
Author :
Arakawa, Hitoshi ; Enokida, Shuichi ; Hayashi, Toyohiro ; Ishikawa, Masashi ; Nobuyama, Eitaku
Author_Institution :
Kyushu Inst. of Technol., Iizuka, Japan
fYear :
2011
fDate :
Nov. 28 2011-Dec. 1 2011
Firstpage :
555
Lastpage :
560
Abstract :
The number of traffic accidents involving cyclists and pedestrians is increasing rapidly in Japan. To clarify the causes of these accidents, it is indispensable to gather information as to what happened in these accidents. This paper introduces the prototype of an Event Data REcorder for a Pedestrian (EDREP) for recording and analyzing traffic accidents similar to the Event Data Recorder (EDR) of a vehicle. The prototype of the EDREP consists of a sensor unit and a controller unit. The sensor unit consists of image sensors, a motion sensor, and a Global Positioning System (GPS) data receiver. The controller unit consists of a personal computer. The sensor unit and the controller unit are connected through a Universal Serial Bus (USB) link. The video image sequences acquired by the EDREP are transferred to a data server through a wireless internet connection in real time using video streaming technology. In this paper, three video camera systems (Type I, II, III) are implemented and assessed. The main difference among the systems is the type of cameras used in the sensor unit. Type I, II and III are equipped with two or more Web cameras, an omni-directional video camera and two or more NTSC video cameras, respectively. This paper also describes the advantages and disadvantages of the three systems based on results obtained by conducting experiments in a real-road environment.
Keywords :
Internet; data recording; image sequences; object recognition; road accidents; traffic engineering computing; video cameras; video recording; video streaming; video surveillance; EDREP; NTSC video cameras; Web cameras; controller unit; data server; event data recorder; image acquisition; omni-directional video camera; pedestrians; traffic accidents analysis; universal serial bus; video image sequences; video streaming; wireless Internet; bicycle; event data recorder; motion sensor; pedestrian; traffic accident; video streaming;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Sensing Technology (ICST), 2011 Fifth International Conference on
Conference_Location :
Palmerston North
ISSN :
2156-8065
Print_ISBN :
978-1-4577-0168-9
Type :
conf
DOI :
10.1109/ICSensT.2011.6137041
Filename :
6137041
Link To Document :
بازگشت