A millimeter-wave radar intersection traffic flow monitoring method and system thereof

Abstract

The invention discloses a millimeter-wave radar intersection traffic flow monitoring method, comprising the following steps: collecting millimeter-wave radar detection information in different scenarios of the intersection; preprocessing the acquired millimeter-wave radar detection information as training samples, constructing a training set, Test set; build a machine learning model based on multi-task convolutional neural network; use the training set and test set to train the machine learning model to obtain appropriate model parameters; collect millimeter-wave radar detection information in real time and input it to the trained machine learning The model classifies, recognizes the vehicle target and quantity; counts the change of the number of vehicles at the intersection within the period, and realizes the monitoring of the traffic flow at the intersection. The invention also discloses a millimeter-wave radar intersection vehicle flow monitoring system. The invention provides a millimeter-wave radar intersection traffic flow monitoring method and its system, which can apply machine learning to the millimeter-wave radar intersection traffic flow monitoring application, and has the advantages of real-time, versatility and high efficiency.

Description

A millimeter wave radar intersection traffic flow monitoring method and system technical field The present invention relates to a kind of millimeter wave radar intersection traffic flow monitoring method and system thereof, belong to intelligent traffic road state. Condition monitoring technology field. Background technique [0002] Millimeter-wave radar has low atmospheric attenuation, better smoke and dust penetration, and is less affected by weather. At the same time, it has high resolution, good directivity, strong anti-interference ability, and good detection performance. It is used for Autonomous driving and intelligent transportation. A system that uses millimeter-wave radar for traffic status monitoring, using traditional radar detection and The statistical method has the following problems: the detection algorithm of millimeter waves for stationary targets is not as good as dynamic targets, and the detection and identification of The false alarm rate is hig...

AI Technical Summary

Problems solved by technology

The system using millimeter wave radar for traffic status monitoring adopts traditional radar detection and statistics methods, and there are the following problems: the detection algorithm of millimeter wave is not as good as the detection algorithm for dynamic targets, and the false alarm rate of detection and recognition is high; The efficiency is limited and the timeliness is not high, so it is difficult to meet the real-time and high-efficiency requirements of traffic flow monitoring at intersections in intelligent transportation; using radar detection, background interference has a great influence. At the same time, different intersections have different backgrounds, and processing parameters must be manually adjusted. poor

Images