Method and system of traffic flow detection based on video identification technology

Abstract

The invention discloses a method and a system of traffic flow detection based on a video identification technology. A reference background image is obtained by adopting a background estimation technology with interval distribution, then information of a moving target in a current frame is obtained according to the result of differential operation between an obtained current frame image by analysis and the reference background image, and the influence of the moving vehicle shadow possibly included in the moving target information is considered to further distinguish the moving shadow and the moving vehicle information to obtain more accurate vehicle flow information. The method comprises the following steps: the road image information is collected in real time and is preprocessed; a reference background image of the road is obtained; the information of a moving target in a current frame image is detected through the differential operation between the current frame image of the road image information and the reference background image, wherein the moving target comprises a moving vehicle or a moving shadow thereof; and the moving vehicle and the moving shadow in the moving target information are distinguished to obtain the traffic flow information.

Description

technical field [0001] The invention relates to a vehicle flow detection method and system, in particular to a vehicle flow detection method and system based on video recognition technology. Background technique [0002] With the rapid development of urban economy, the demand for road traffic is increasing rapidly, and traffic congestion and congestion are becoming more and more serious. It directly affects the efficiency and safety of urban transportation, aggravates the environmental pollution of the city, and becomes a restrictive factor for urban economic construction. For this reason, how to implement modern control and guidance on the existing road traffic network has become particularly important. [0003] However, traffic control and guidance must be based on real-time and accurate traffic flow statistics, and manual statistics in the past can no longer meet the needs. As a result, various traffic parameter detection methods have emerged, mainly including ultrasoni...

AI Technical Summary

Problems solved by technology

However, each method has its disadvantages to be overcome. For example, the ultrasonic detection accuracy is not high, it is easily affected by vehicles and pedestrians, and the detection distance is short; the infrared detection is affected by the heat source of the vehicle itself, and its anti-noise ability is low. The accuracy is low; the detection accuracy of the ring induction coil is high, but it needs to be installed in the civil structure of the road surface, which will damage the road surface and make construction and installation inconvenient

Images