Traffic flow statistical method in intelligent traffic

Abstract

The invention relates to a traffic flow statistics method in intelligent traffic. In the prior art, the accuracy of vehicle identification is insufficient, and the efficiency is not high. According tothe method, an SSD and a ResNet in a neural network are adopted, and the method is established with a traditional target tracking CamShift algorithm and comprises a vehicle detection method and a vehicle tracking method. According to the vehicle detection method, firstly, an SSD network is established, the SSD obtains a plurality of feature maps of different sizes, different aspect ratios are adopted for default frames on the same feature layer, the robustness of the default frames to the shape of an object is enhanced, and the position and the target type are regressed while the SSD trainingis carried out. According to the vehicle tracking method, a continuous self-adaptive expected movement algorithm is adopted to track a vehicle identified by a first frame of a single-time movement visual network detector. According to the method, the vehicles can be detected more accurately for statistics, the vehicles can be prevented from being lost through tracking to be counted again, and theobtained traffic flow data are more accurate.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Benefits of technology

This technology helps count how many people are driving on roads faster than they actually see or react accordingly. It uses advanced sensors that track their movements over time to make it easier to identify specific areas where there may be potential hazards like cars or pedestrians. By analyzing these movement patterns, we aimed at finding ways to improve road safety measures such as speed limit signs or lane markings.

Problems solved by technology

Technological Problem: Current Vehicle Detection (Vehicles Detect) Methodologies have limitations such as slow response times due to factors like noise interference caused by surrounding buildings, difficulty recognizing different types of cars, lack of environmental awareness during operation, etc., making them difficult to achieve accurate results while maintaining efficient processing capabilities over varying conditions.

Images