Statistical method for traffic flow based on moving vehicle detection

Abstract

The invention discloses a statistical method for the traffic flow based on moving vehicle detection, and relates to the field of computer vision. The statistical method comprises the steps of firstly performing image preprocessing by using a multi-scale morphological operator, then detecting a vehicle target by combining background real-time update, background difference and the edge gradient difference, and thus acquiring a binary image of the vehicle target; and finally performing single-lane or multi-lane traffic flow statistics through dual thresholds and state transition of a lane detection coil according the binary image. According to the method, the traffic flow of a single lane or multiple lanes can be worked out effectively and quickly, traffic lights can be enabled to change the time of red and green lights according to the current traffic flow, the utilization rate of urban transportation is improved, and a problem of urban vehicle congestion is improved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a traffic flow statistics method based on moving vehicle detection. Background technique [0002] With the continuous development of artificial intelligence, computer vision and hardware technology, video image processing technology has been widely used in intelligent transportation systems (ITS). In recent years, with the promotion of highway video surveillance, image processing methods have begun to be applied to the field of traffic analysis, including traffic incident detection, traffic queue monitoring, vehicle type recognition, vehicle classification, traffic flow statistics, etc. [0003] The significance of real-time traffic flow statistics is that the quickly counted traffic flow can provide effective data for signal light control, so that traffic signals can change the length of traffic lights in real time according to the size of the current traffic flow, thereby improvi...

AI Technical Summary

Problems solved by technology

Wherein, the adjacent frame difference method is not sensitive to illumination, is very suitable for a dynamically changing environment, and has simple calculation, fast detection speed, accurate vehicle positioning, and is suitable for application environments with high real-time requirements; but it cannot detect static or The moving speed of the object is too slow, so this method fails when there is congestion at the intersection, and it is easy to cause a large hole in the overlapping part of the target. In severe cases, the segmentation results will be disconnected, resulting in multiple or wrong detection of vehicles

The background difference method is suitable for the case where the camera is still, and the calculation speed is fast, but it is sensitive to the lighting conditions, large-area motion and noise in the scene

The optical flow method can also detect independent moving targets when the camera is moving, but its disadvantages are that it is susceptible to noise interference and has a large amount of calculation. If there is no specific hardware device, it cannot be applied to the real-time monitoring of full-frame video streams. deal with

Images