主催: 一般社団法人 日本機械学会
会議名: Dynamics and Design Conference 2017
開催日: 2017/08/29 - 2017/09/01
Recently, traffic congestions caused by increase in the number of vehicle possession and complicatedness of traffic system have been serious social problems. In order to solve these problems, a lot of attempts have been carried out from various aspects including employing new type of traffic signals with fuzzy control system or neural network system. Several types of simulation systems have been considered as an effective technique to determine how to reduce the traffic load. However, the current simulation system has several problems, which cannot model after on real traffic system. To obtain more reliable simulation results, we need a simulation model that is able to control more accurately frequency of car appearing, type of car, speed, etc. In this study, a traffic flow measurement system has been developed to extract traffic flow data by analyzing images near the intersection from the fixed point camera. The measurement system has been developed by program language C++ and OpenCV library and performance of the measurement system was measured based on recognition rate of the number of cars passing the intersection, the type of cars and driving means speed of passing between two intersections. The average moving speed and whether route change is happening between intersections are measured and presumed by identifying the same vehicle from image data of two locations. It could be confirmed that this traffic measurement system might be available to traffic census and also input data into the simulation system through this study.