Traffic flow statistics and vehicle type classification method and device

Abstract

The invention belongs to the field of data, and relates to intelligent traffic, in particular to a traffic flow statistics and vehicle type classification method and device. The method comprises the following steps: acquiring a traffic flow video, setting a rectangular identification frame in a video image sequence, and taking a frame without vehicle passing in the identification frame as a background frame; obtaining a differential image of the current frame and the background frame; storing the gray scale average value in a gray scale feature array; calculating a vehicle entry threshold value and a vehicle driving-out threshold value in the statistical time window; comparing the gray average value with the two thresholds respectively, and if the gray average value is greater than the entry threshold, adding 1 to the count of a vehicle counter and freezing the counter; if the gray average value is smaller than the driving-out threshold value, enabling a counter; and finally obtainingtraffic flow data; if the time duration is greater than the vehicle entry threshold, adding 1 to the counter, and resetting the timer; if the time duration is smaller than the vehicle driving-out threshold , freezing the timer; storing the vehicle model characteristic data into a vehicle model characteristic array; and calculating classification thresholds, comparing the vehicle model characteristic values with the classification thresholds, and finally obtaining a vehicle model classification result.

Description

technical field [0001] The invention belongs to the field of data and relates to intelligent transportation, in particular to a method and device for vehicle flow statistics and vehicle classification. Background technique [0002] With the continuous development of cities, people drive more and more frequently, and road congestion has become an urgent problem to be solved, which puts forward higher requirements for urban road construction and traffic management. Smart transportation uses big data to analyze the driving characteristics of the road (vehicle flow and model characteristics at each time period), predict and guide traffic, and provide guidance for urban road construction. [0003] Obtaining road traffic characteristic data is the basis for traffic road analysis, management and construction. The traditional method is to manually count and classify specific intersections. This method has low efficiency, high cost, less data, and large errors. With the development ...

AI Technical Summary

Problems solved by technology

However, the disadvantage of these methods is that the feature extraction is too complicated to reduce the processing efficiency, and the fixed threshold reduces the accuracy.

Images