Intelligent traffic light control method and system

Abstract

The invention relates to an intelligent traffic light control method. The method comprises: video image information of a road vehicle is obtained; pretreatment is carried out on the obtained video image information and a pre-training network model is constructed, a training network is built based on the pre-training network model, pre training is carried out on the built pre-training network, the training network is initialized by using pre-trained network parameters and the training network is trained by using a constructed data set until a vehicle number at a corresponding lane in a picture is detected in real time; and on the basis of original passing time, the detected vehicle number at the corresponding lane used as a parameter is substituted into a crossroad traffic light passing standard and the crossroad traffic light passing time is adjusted in real time. In addition, the invention also relates to an intelligent traffic light control system. According to the method and system provided by the invention, the green light passing time of the crossroad traffic light can be adjusted in real time based on vehicle recognition, so that road congestion can be reduced, the traffic accident can be avoided, and the passing efficiency can be enhanced.

Description

technical field [0001] The invention relates to an intelligent traffic signal lamp control method and system. Background technique [0002] The current traffic signal light control system attempts to control the traffic signal lights by identifying vehicles on the road in various ways. It is roughly divided into three types: magnetic frequency detection, wave frequency detection and video detection. The basic technologies relied on are different and each has its own advantages. [0003] At present, the most widely used at home and abroad is magnetic frequency detection, that is, the vehicle detector is a loop coil, which is buried under the road surface to detect the traffic volume above the road, the road occupancy ratio and the speed of vehicles passing, but the laying of the equipment needs to be destroyed The ground is inconvenient to maintain and upgrade in the later stage, and the scalability is poor; the wave frequency detection is based on microwave, ultrasonic or ...

AI Technical Summary

Problems solved by technology

[0003] At present, the most widely used at home and abroad is magnetic frequency detection, that is, the vehicle detector is a loop coil, which is buried under the road surface to detect the traffic volume above the road, the road occupancy ratio and the speed of vehicles passing, but the laying of the equipment needs to be destroyed The ground is inconvenient to maintain and upgrade in the later stage, and the scalability is poor; the wave frequency detection is based on microwave, ultrasonic or infrared waves, etc., to detect the electromagnetic waves emitted by the vehicle, and the accuracy is acceptable, but the cost is high; with deep learning Wide application in the field of image recognition, image processing based on video detection, and vehicle recognition has become a hot spot in recent years, but has not yet been fully applied in practice

[0004] In short, the above traditional detection methods for vehicle identification and then control of traffic lights cannot achieve satisfactory results in terms of efficiency and accuracy.

Images