Study of abnormal incident detection aimed at automatic wide-area traffic flow monitoring.

Author(s)
Oikawa, K. Kaneko, Y. & Matano, M.
Year
Abstract

Since 1994, researchers at the Metropolitan Expressway Public Corporation (MEPC) have been experimenting and examining Abnormal Incident Detection Systems which employ image processing technology. Because the influences of a vehicular accident on the following vehicles are so great, they are looking into the introduction of such systems into Metropolitan Expressway with top priority given to the tunnel sections of the expressway. In 1997, they introduced an Abnormal Incident Detection System into the Chiyoda, Kasumigaseki, Kuko-Kita, and Tokyo-Ko Tunnels, respectively, on a trial basis, by making the best use of their existing closed circuit television (CCTV) cameras. As regards the tunnel sections, as well as bright roadway sections in which the number of accidents is relatively great, they make it a rule to adopt a constant processing method for the system to be introduced, in which one image processing unit (IPU) constantly handles images from only one camera. It can be said that the constant processing method is ideal in terms of speed and reliability in the detection of an abnormal incident. However, due to the fact that they have already installed nearly 1,000 CCTV cameras throughout the road network of the expressway and in view of their future plan of introducing the system into all the routes of the expressway, an enormous cost of system introduction is nothing but a headache to the public corporation. So, they are looking into the adoption of a switchover processing method in which two or more cameras are handled by only one IPU. This switchover processing method has such a demerit that it has some delay in detecting an incident as compared with the constant processing method. To minimize this delay in the incident detection time, they are looking into the adoption of a variable cycle method. (In reference to this variable cycle method, they call the former method a fixed cycle method.) The researchers conducted verification testing on both the variable and fixed cycle methods, analyzed the results of testing comparatively and confirmed the superiority of the variable cycle method to the fixed cycle method as reported in this paper.

Request publication

3 + 0 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.

Publication

Library number
C 33162 (In: C 26095 CD-ROM) /80 / ITRD E828466
Source

In: ITS - Transforming the future : proceedings of the 8th World Congress on Intelligent Transportation Systems ITS, Sydney, Australia, 30 September - 4 October 2001, Pp

Our collection

This publication is one of our other publications, and part of our extensive collection of road safety literature, that also includes the SWOV publications.