Traffic Analysis of Important Road Junctions Based on Traffic Flow Indicators

摘要

Intelligent transportation systems can be used as an indicator of urban sophistication. Therefore, many countries devote themselves to developing an intelligent transportation system. The construction of intelligent transport system in Taiwan is still in infancy. The allocation of traffic signal lights requires a lot of manpower and time to observe the site. Based on collected information, data integrity was inadequate due to human negligence. The traffic congestion problems caused by the mistake can't be improved. In this article, the traffic flow based on the intersection traffic flow indicators analysis is proposed. This method is based on the relationship between data to judge, to avoid the problem of empirical judgment error without human intervention. Traffic flow are collected by the vehicle detector to improve data integrity. Using the indicator for road junction association analysis can find the association rules between road junctions. Finally, a deep neural network is used as its classifier, so the accuracy is over 95%.