A discretized traffic state discrimination method based on deep learning

Abstract

A discretized traffic state discrimination method based on deep learning belongs to the technical field of traffic state discrimination. The method includes the following steps: step a, using the monitoring equipment installed in each direction of the intersection to collect the traffic status of the intersection at multiple different moments, and correspondingly obtain multiple sets of traffic status information; step b, according to the discretization coding process The information is described by discretization encoding respectively, corresponding to the discretization encoding data of the intersection at different times; step c, constructing a traffic state discrimination model based on deep learning according to the discretization encoding data obtained in step b; step d, using step c The constructed traffic state discriminant model is used to discriminate the traffic state. Through this discretized traffic state discrimination method based on deep learning, the traffic state can be comprehensively and truly described, no need for experts to participate in the selection of traffic state features, and the construction of the traffic state discrimination model can be realized automatically under semi-supervision.

Description

technical field [0001] A discretized traffic state discrimination method based on deep learning belongs to the technical field of traffic state discrimination. Background technique [0002] With the acceleration of social urbanization, large and medium-sized cities around the world are facing the pressure of traffic congestion. Similarly, the number of household cars in my country has continued to grow rapidly in recent years, and urban traffic congestion has become the focus of discussion on public travel. To solve the problem of traffic congestion, there are usually three solutions: (1) Control vehicle travel through administrative means, for example, some cities in my country implement The odd and even number restriction system, the scheme is simple and easy to implement, but it will lead to the embarrassing situation of "cars are not allowed"; (2) expanding transportation facilities and building new traffic roads, this scheme requires a lot of manpower, material resources...

AI Technical Summary

Problems solved by technology

[0005] (2) The traffic state is a complex combination of factors in the real traffic environment, and the way of manually setting the state characteristics will lose relevant valuable information

[0006] (3) The established traffic state discrimination model has poor adaptability when applied to new traffic environments

Images