Vehicle flow detection algorithm based on video image processing

Abstract

The invention relates to a vehicle flow detection algorithm based on video image processing. The concrete algorithm comprises the following steps: a. continuous frame sequence images are inputted; b. image graying and smoothing are performed; c. a gray scale histogram and a gray scale mean value of each pixel point of an image frame sequence are counted; d. a background model and a threshold plane are established; e. frame inputting and background difference are performed, operation target regions are extracted, an OTSU is performed after the threshold plane is established, and a differential image threshold T is calculated; f. shadow detection based on color and gradient is performed, and binary and morphological filtering processing is performed after the threshold T is calculated; g. information of a moving vehicle of which shadow is eliminated is acquired; and h. vehicle flow detection based on coils is performed. The vehicle flow detection algorithm based on video image processing is simple in principle, convenient to calculate and more accurate in extraction of vehicle flow data so that a function of counting of vehicle flow predetermined by a system can be greatly realized via large amount of experiments and simulations.

Description

technical field [0001] The invention relates to the field of video images, in particular to a traffic detection algorithm based on video image processing. Background technique [0002] In today's society, the high-speed economic development has made people's demand for materials continue to increase, and the number of motor vehicles, especially private cars, has increased rapidly. Although road construction is also improving day by day, including effective measures such as widening roads and building viaducts, but after all, it cannot solve the fundamental problem. A series of problems such as congestion and accidents caused by more and more traffic flow have caused more and more serious problems in people's daily life. more serious impact. Therefore, the reasonable evacuation control of traffic flow has become the focus of research in this field. In the era of increasing information technology development and gradual improvement and updating, intelligent transportation sy...

AI Technical Summary

Problems solved by technology

This method has mature technology and reasonable price, but because its detection accuracy is affected by environmental problems and the range of action of the coil, it is only widely used in highway, bridge and tunnel lighting environments, and is not suitable for complex urban traffic environments

In addition, this method will cause damage to the ground when the coil is buried, hinder the traffic of the installation and maintenance team, and cause higher maintenance costs in the later period

[0005] The wave-frequency vehicle detection method is to transmit electromagnetic waves to the vehicle through wave-frequency signals such as infrared rays, microwaves, and ultrasonic waves and receive the signals reflected by the vehicle, thereby calculating the corresponding traffic parameters such as traffic flow and average speed. The method is mostly For the hanging type, the detector requires a space of 3m from the nearest vehicle during detection, and must meet a certain installation height requirement, so the installation on bridges, overpasses and viaducts will be limited, and the system is relatively expensive

Images