Quality detection method of city-level traffic video detection equipment

Abstract

The invention discloses a quality detection method for city-level traffic video detection equipment. The method comprises the following steps: a. Constructing the network topology structure of each video detection device in the road network, and obtaining a network topology table; b. Constructing a randomness model of road section travel confidence time according to the network topology table; c. Identifying the video detection device The equipment detected by S detects correct passing data and wrong data in the total amount N, and calculates the false detection rate of video detection equipment S; traverses all video detection equipment, and obtains the false detection rate of all video detection equipment; d. According to the road section The travel confidence time is used to randomly simulate the missing data, define the stability index of the passing traffic volume in the road section, and calculate the missed detection rate of the video detection equipment in the network topology table. The invention has the characteristics of improving traffic data quality, reducing workload and improving data description accuracy, and is the basis for intelligent traffic platform construction.

Description

technical field [0001] The invention relates to the technical field of data quality monitoring, in particular to a quality detection method for city-level traffic video detection equipment. Background technique [0002] Most cities in China have experienced several years of intelligent transportation construction, installed a large number of traffic data collection equipment, and collected massive traffic data. However, when the traffic management department wants to use big data technology to alleviate urban traffic congestion, it is very difficult. The main reason is that during the equipment assembly process, due to the lens quality, assembly process and other reasons, the accuracy rate will drop slightly. Interferenced by factors such as light, equipment installation angle, traffic density, license plate suspension method, etc., the collected data inevitably has noise. Data sets with a large number of errors cannot support the effective use of smart transportation syste...

AI Technical Summary

Problems solved by technology

However, when the traffic management department wants to use big data technology to alleviate urban traffic congestion, it is very difficult. The main reason is that during the equipment assembly process, due to the lens quality, assembly process and other reasons, the accuracy rate will drop slightly. Interferenced by factors such as light, equipment installation angle, traffic density, license plate suspension method, etc., the collected data inevitably has noise

Datasets with a large number of errors cannot support the effective use of intelligent transportation systems

And at present, it only monitors the power failure of the equipment, and has not tested the indicators such as the error rate and missed detection rate of the equipment.

Images